Nelson Advisors Big Questions in HealthTech Series: Is the LLM an interface or the decision-maker? The structural evolution of large language models has shattered the primitive paradigm of the conversational chatbot, prompting a fundamental re-evaluation of system design. Modern software engineering faces a critical dichotomy: should the language model serve as an intuitive cognitive interface, a translation layer interpreting human intent into structured machine directives, or should it function as a sovereign, system-level decision-maker, capable of autonomous planning and resource allocation? Resolving this question requires dissecting the mathematical limits of autoregressive transformers, examining the structural parallels between generative systems and operating system kernels, and analysing emerging neuro-symbolic frameworks. This structural debate coincides with the emergence of what industry architects term "Software 3.0". In traditional software paradigms, the application's logic is entirely deterministic and hard-coded by developers who map every execution pathway. In contrast, the AI-native software model reorganises this hierarchy by positioning the language model as a universal controller. In this architectural pattern, traditional application boundaries dissolve; individual applications are recast as a unified registry of plug-ins orchestrated by a central model. Rather than requiring human operators to manually coordinate tasks across disconnected software silos, the model-as-controller receives unstructured natural language, reasons over intent and composes sequential actions across diverse plugin APIs. Yet, this flexibility introduces severe logical non-determinism, forcing system designers to rigorously map the operational boundaries where a model excels as a semantic interface and where it fails as an unconstrained decision-maker. The LLM as Cognitive Interface: Semantic Translation, Intent Parsing and AutoFormalisation To evaluate the role of the model as an interface, system engineers must analyse how it bridges the semantic gap between high-dimensional human intent and low-level, deterministic computing protocols. Autoregressive language models excel at processing language because human communication is built on low-dimensional, compressible patterns. Rather than acting as precise databases, these networks function as vast, non-veridical memories that probabilistically reconstruct outputs token by token, a process characterised as approximate retrieval. This probabilistic quality makes them exceptionally flexible as interfaces; they parse human commands, analyse user intent and translate unstructured requirements into domain-specific languages, SQL queries, or API parameters. This translation paradigm is highly effective in infrastructure orchestration and data analytics. For instance, in cluster scheduling, Kubernetes schedulers have been successfully augmented with language-model-based intent analysers. This architecture leverages the model strictly as a translation layer that interprets unstructured natural language annotations representing soft-affinity preferences (e.g., placing workloads near specific data sources or on lightly loaded nodes) and parses them into low-level scheduling directives, achieving an empirical accuracy greater than 95%. Similarly, in product analytics, systems maintain a deterministic data ingestion pipeline while employing the language model exclusively at query-time to translate natural language questions into ClickHouse SQL. By isolating the model from the database write-path and enforcing strict schema validation on the generated queries, the database remains a stable source of truth while benefiting from an intuitive, conversational interface. Beyond query translation, the model-as-interface plays a critical role in "AutoFormalisation". This refers to the automated translation of informal natural language descriptions into formalised symbolic representations, such as first-order logic, mathematical proofs, or Planning Domain Definition Language (PDDL) models, which can then be processed by deterministic symbolic solvers. This hybrid design allows systems to achieve "epistemic humility". As demonstrated by policy and development research systems like AVA (built on World Bank reports), the integration of structured retrieval pipelines with language-model interfaces allows the system to enforce strict citation verifiability and "reasoned abstention", the capability of the interface to decline answering queries when the underlying grounded data is insufficient, preventing the hallucinations common in unconstrained generative models. System-Level Implementations: The Large Language Model as an Operating System Kernel While the translation layer paradigm frames the model as an interface, an alternative system-level framework likens the Large Language Model to an operating system kernel. Pioneered by researchers and formalised in Artificial Intelligent Operating System (AIOS) literature, this perspective treats the model as a core computational processor rather than a simple text-generation utility. Within this system architecture, traditional hardware abstractions find direct cognitive equivalents. Traditional Operating System Component AIOS / LLM Operating System Equivalent System-Level Operational Role Central Processing Unit (CPU) / Kernel Large Language Model Core Executes core cognitive operations, processes intent, and arbitrates system actions Random-Access Memory (RAM) Context Window Volatile working memory; handles immediate context selection and active data processing Hard Disk Storage / File System External Storage / Retrieval-Augmented Vector Stores Persistent long-term storage of documents, logs, and historical context Peripheral Devices Hardware Tools / Actuators Connects the system to the physical world (e.g., robotic arms, sensors, on-vehicle cameras) Programming Libraries / APIs Software Tools / SDK Plug-ins Extends core capabilities to execute arithmetic, code generation, or database writes User Commands / Shell Executables Natural Language Prompts Initiates system actions and configures operating environments FIFO Schedulers / Thread Queues Reasoning Loops / System Calls Coordinates concurrent agent executions, optimises prompt pipelines, and prevents CUDA crashes In this AIOS framework, resource management and scheduling diverge fundamentally from traditional computing. Instead of managing raw hardware clock cycles and memory addresses, the AIOS kernel manages context windows and tool tokens, orchestrating execution through a continuous reasoning loop rather than a FIFO queue. Standard agent orchestrators (such as early implementations of Autogen or Langchain) run directly on host-level environments and execute model API calls via brute-force trial-and-error. Under concurrent execution workloads, this approach causes GPU memory saturation, triggering CUDA exceptions that force expensive tensor deallocations and multiple retry cycles. The AIOS kernel layer resolves this resource bottleneck by isolating agent applications from direct system-level resources. The kernel decomposes incoming agent requests into standardised system calls (syscalls) which are systematically scheduled across separate storage, memory, and tool managers. By orchestrating syscall execution across a unified interface, the scheduler prevents concurrent agent requests from flooding the model, resulting in up to a x2.1times increase in execution speed for serving concurrent agents. This architecture enables a new AIOS-Agent ecosystem, where specialised Agent Applications (AAPs), such as trip planners, financial advisors, or medical consulting agents, are deployed as OS-native applications that leverage the intelligent scheduling and tool-execution capabilities of the underlying kernel. The Physical Realisation: Dedicated Coprocessor Hardware for Local Inference The conceptualisation of the model as an operating system kernel is driving a corresponding shift in physical computer architecture: the emergence of dedicated neural processing units (NPUs) acting as specialised language model coprocessors. Hardware designers are developing application-specific integrated circuits (ASICs) optimised to run specific open-weight models locally. A prime example is Rockchip's RK182X coprocessor, which incorporates ultra-high-bandwidth memory to run models like Qwen 2.5 7B at $50 { tokens per second (TPS)}for token decoding and up to $800 { TPS} for prompt processing. To understand the efficiency of these hardware designs, one must analyse the mathematical divergence between the two primary phases of inference: Prompt Processing (PP) and Token Generation (TG). Operational Phase Computational Bottleneck Hardware Resource Dependency Local Hardware Optimization Mechanics Prompt Processing (PP) Compute-Limited Parallel Tensor Core Execution (FLOPs) Loads the model weights into cache once; processes all context tokens in parallel across highly parallelised, low-precision tensor cores Token Generation (TG) Memory-Bandwidth Limited Memory Access Speed and Bus Bandwidth (GB/s) Must sequentially load every single layer of network weights from RAM to generate a single token, bottlenecked by system memory bus speeds Traditional NPUs typically utilise the host system's slow main memory, rendering them ineffective during the memory-bandwidth-bound token generation phase. Coprocessors like the RK182X bypass this constraint by pairing the NPU with dedicated, high-speed on-chip memory of ultra high-bandwidth LPDDR), allowing Q4-quantised models to load weights almost instantaneously near the arithmetic logic units (ALUs). By offloading matrix multiplications to low-power ASICs that consume up to 90% less power than standard GPU setups, devices can maintain large context windows and execute background reasoning routines locally on edge devices without thermal throttling or remote server dependency. Mathematical and Empirical Limits of Autonomous Decision-Making Despite the appeal of the "model as operating system" paradigm, treating autoregressive language models as sovereign, autonomous decision-makers introduces severe mathematical and logical failures. By definition, autoregressive transformers predict the next token sequentially, running in constant computational time per step. However, mathematical reasoning, planning, and logical verification are often NP-hard or semi-decidable problems that require variable-time combinatorial search. A system constrained to constant-time state transitions cannot perform principled logical deduction; it can only simulate reasoning by retrieving and recombining patterns from its training corpora. Because of this limitation, the performance boundaries of these systems are highly irregular. While a model may answer complex, Olympiad-level questions that resemble patterns in its pre-training data, it can simultaneously fail at basic arithmetic operations. Rigorous evaluations on classical planning benchmarks, such as the blocks-world problems in the International Planning Competition (IPC), highlight this deficit. When evaluated in an autonomous planning mode on platforms like PlanBench, even advanced models like GPT-4 only generate valid, executable plans approximately 12% of the time. Evaluation Benchmark Primary Reasoning Modality Tested Model Performance Characteristics Underling Computational Failure PlanBench (IPC Blocks-World) Combinatorial Search and Subgoal Sequencing GPT-4 achieves ~12% autonomous plan correctness; performance drops near zero under term obfuscation Relies on approximate retrieval of memorised plan structures rather than active logical sequencing ACPBench Hard Generative Action, Change, and Multi-Step Planning Frontier models, including early deliberate reasoning models, score below 65% across most tasks Struggle to resolve causal action-precondition dependencies in a generative format AIME 2024 Advanced Mathematical and Deductive Reasoning Standard GPT-4o scores ~12% pass@1; deliberate reasoning models (o1) achieve up to 74% pass@1 Standard models fail on complex multi-step chains; reasoning models succeed by scaling inference-time search tokens OPT-BENCH Continuous Optimization vs. Discrete Combinatorial Reasoning Strong on continuous inductive tuning (ML hyperparameter optimization); poor on NP-hard discrete search Lacks internal execution verification models to navigate brittle, discrete search spaces The reliance of autoregressive models on memorised structures is further demonstrated by domain obfuscation tests. When standard planning terms (e.g., "stack," "unstack," "block") are replaced with random, non-semantic strings, a change that does not affect deterministic symbolic solvers, the model's planning accuracy collapses entirely. Furthermore, the common architectural assumption that models can self-correct through iterative evaluation loops is flawed. Because autoregressive models cannot reliably verify their own solutions, iterative self-critique often degrades plan quality. Lacking an internal logical model, the system frequently abandons correct intermediate solutions and replaces them with incorrect alternatives, leading to cascading errors. Yann LeCun's Alternative: World Models and Non-Generative Architectures This computational deficit forms the basis of Yann LeCun's critique of the generative AI paradigm. LeCun argues that language is a low-dimensional, highly compressed representation of human intelligence, whereas the real world is continuous, noisy, high-dimensional, and sensory-rich. Autoregressive models operate purely in this discrete textual space, making next-word predictions without developing an internal, causal model of physical reality. While a model can statistically associate terms like "glass" and "shatter," it lacks the fundamental physical common sense possessed by a house cat, which can predict the gravitational and mechanical consequences of its actions in the physical world. To achieve true System-2 planning and reasoning, LeCun advocates for objective-driven AI built on Joint Embedding Predictive Architectures (JEPA) rather than generative models. Instead of predicting raw pixels or words, JEPA-style architectures learn to predict abstract, high-level representations of the world, filtering out irrelevant noise (such as the movement of leaves on a tree) to focus on predictable, causally relevant information. These models serve as predictive world models, allowing an AI agent to simulate the outcomes of its actions internally and optimise plans before executing them in the physical environment. This paradigm enables hierarchical planning, which is the capability to plan actions at varying levels of abstraction. When planning a trip from New York to Paris, a human does not plan the precise sequence of muscle movements required to walk; instead, the trip is decomposed into higher-level logical chunks (e.g., drive to the airport, board the plane, land in Paris). Modern Vision-Language World Models (VLWM) like Virgo attempt to realize this by learning an action policy (representing reactive System-1 behavior) alongside a predictive dynamics model (representing reflective System-2 behavior). By compressing sensory video data into structured abstractions (e.g., a "Tree of Captions") and utilising self-supervised critics to evaluate hypothetical future states, these models allow agents to perform internal trial-and-error to find cost-minimising action plans, bringing a level of physical grounding and logical consistency that pure language models cannot achieve. Hybrid Architectures: Neuro-Symbolic Synthesis and Deliberate Inference To balance the flexibility of language models with the logical precision required for enterprise execution, system architects categorise application capabilities along a spectrum of six distinct levels of autonomy. This classification maps the boundary where human-defined, deterministic constraints end and non-deterministic model actions begin. Autonomy Level Core Technical Abstraction Control Flow Architecture Primary Security & Operational Risk Level 1: Code Traditional Software Explicit, hard-coded, deterministic logic written by developers Low; standard software testing and compiler constraints apply Level 2: LLM Call Isolated Call Model outputs a single prediction for a predefined, isolated step Minimal; easily validated by standard text extraction rules Level 3: Chain Fixed Pipeline Output of step $n$ feeds directly as input to step $n+1$ in a static sequence Low; data-flow boundaries are predictable and sandboxed Level 4: Router Acyclic Selection Model evaluates context and selects from a predefined set of acyclic pathways Moderate; requires strict validation of the selected execution path Level 5: State Machine Cyclic Workflow Model dynamically decides the next execution step in a workflow that includes loops High; workflows can loop indefinitely, requiring runtime execution budgets Level 6: Autonomous Unconstrained Agent Model independently determines goals, selects tools, and executes actions Severe; requires persistent sandboxing and emergency kill-switches To safely deploy systems at Levels 5 and 6, modern software engineering rejects simple pipelines in favour of neuro-symbolic designs that couple the semantic capabilities of models with external, deterministic verification engines. Three paradigms illustrate this integration: The LLM-Modulo Framework This framework establishes a tight, bi-directional loop where the model functions as an approximate proposal generator or domain translator. The actual validation of plans is managed by a bank of external critics. Hard Critics, such as classical PDDL planners or verification tools like VAL, evaluate plans for causal correctness, physical feasibility, and resource constraints. Soft Critics, often driven by separate, specialized vision-language models, evaluate abstract qualities such as style, conformance, and user preferences. If a hard critic identifies a logical error, it generates precise symbolic feedback. The model consumes this feedback and generates a refined candidate plan. Crucially, expert humans are excluded from this inner planning loop, interacting only in the outer loop to define domain specifications and preference models, preventing cognitive fatigue and the "Clever Hans" effect. Agentic Fast-Slow Planning (AFSP) Inspired by dual-process cognition, this framework decouples perception, reasoning, and control across distinct timescales to ensure physical safety in real-time systems. The system is split into two bridges: Perception2Decision: A local, edge-based vision-language model topology detector compresses raw physical inputs into compact egocentric topology graphs, which are then transmitted to a cloud-based language model to generate high-level symbolic driving directives. This reduces bandwidth and latency while maintaining operational interpretability. Decision2Trajectory: High-level symbolic directives are converted into physically feasible paths using a classical search algorithm that embeds soft costs derived from the model's directives into geometric trajectory optimization. An online Agentic Refinement Module monitors execution and dynamically tunes hyper-parameters using feedback and memory, ensuring the system adapts to environmental changes without bypassing classical safety bounds. Deliberate Inference Reasoning Models Models such as OpenAI o1, DeepSeek-R1, and QwQ attempt to internalise System-2 deliberate thought during inference through test-time compute scaling. Instead of predicting the next token instantly, these models are trained via reinforcement learning algorithms, such as Group Relative Policy Optimisation (GRPO), to generate extensive internal scratchpads before committing to a final, visible answer. GRPO reduces training overhead by dropping the traditional value-function model, estimating baseline rewards across a group of sampled answers to penalise logical inconsistencies and reward correct, verifiable solutions. Through this reinforcement loop, reasoning models naturally learn to explore alternative pathways, check intermediate steps, and backtrack when an error is detected. While this approach significantly improves performance on complex mathematical and coding benchmarks, it introduces an "overthinking" efficiency bottleneck, where models consume excessive tokens and compute resources solving simple tasks that could be resolved instantly with minimal token generation. Security Implications of the Autonomy Shift: Vulnerability Analysis and Threat Mitigation As systems transition from using models as cognitive interfaces to deploying them as active decision-makers with tool-execution privileges, the primary threat vector shifts from content moderation to critical system-level vulnerabilities. In traditional computing, code injection exploits deterministic parsing bugs to execute compiled binaries. In agentic computing, prompt injection exploits the model's inability to structurally separate trusted instruction sets from untrusted data inputs. In agentic systems, prompts serve as non-deterministic programs written in natural language. Direct prompt injections occur when a user actively inputs commands to override safety bounds. Far more dangerous, however, are indirect prompt injections, where malicious instructions are hidden inside external resources parsed by the agent at runtime, such as emails, PDF documents, or web pages. When an agent is granted tool access, a successful indirect prompt injection can hijack its goals, manipulating it into performing unauthorised API calls, executing production database writes, or exfiltrating sensitive company files. This vulnerability is highly evident in modern, high-privilege agentic coding environments and editors like Cursor. Designed with system-level access to execute terminal commands, edit local files, and interact with external systems, these agents can be targeted by poisoning external development resources, such as configuration files, repository documentation, or Model Context Protocol (MCP) server definitions. When the agentic editor processes these poisoned resources during standard development routines, the embedded instructions hijack its execution flow. This converts the agent into an attacker's terminal, enabling remote command execution, code injection into production repositories, and local machine compromise without any human clicking a link or running an exploit binary. To defend agentic environments against these vulnerabilities, security teams must deploy multi-layered defense frameworks like MAESTRO. First, developers must enforce the principle of least-privilege, ensuring that agent credentials are sandboxed and restricted to read-only access where possible. Second, systems must implement cryptographic prompt signing workflows, where trusted instructions are signed with a corporate cryptographic key. When instructions pass through multiple agents, the execution kernel verifies the signatures; if untrusted user data attempts to append system-level directives, the signature validation fails, and the execution is terminated. Finally, designers must place human-in-the-loop approval gates before any high-stakes, irreversible tool executions are processed, establishing an administrative barrier against runaway or hallucinatory actions. Conclusions and Strategic Recommendations The technical evidence establishes that the Large Language Model is fundamentally suited to serve as a cognitive interface and approximate proposal engine, rather than a sovereign, unconstrained decision-maker. While operating-system-level integrations and test-time compute scaling have expanded the capability of these models to navigate complex environments, their underlying probabilistic nature prevents them from providing the absolute logical guarantees required for high-risk system automation. When deployed as interfaces, language models act as powerful cognitive orthotics, translating human intent into actionable configurations and helping to bridge complex semantic gaps. However, the sovereign decision-making authority must reside in model-based, symbolic, or human-controlled verification layers. Modern hybrid frameworks like the LLM-Modulo and Agentic Fast-Slow architectures demonstrate that the future of resilient system design does not rely on scaling models indefinitely. Instead, it lies in the structured integration of neural and symbolic components, leveraging the semantic versatility of language models to map intent, while relying on formal verification systems to execute and validate actions. By enforcing these architectural boundaries, engineers can safely build intelligent, highly adaptive systems that preserve security, predictability, and logical soundness in production environments. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

This Week in European MedTech and HealthTech: 3rd July 2026 The European HealthTech landscape is experiencing a definitive shift away from "growth-at-all-costs" toward highly disciplined, clinical, and regulatory-compliant solutions. This week’s major developments emphasise interoperability, operational efficiency, and deep-tech medical devices. 1. Deep-Tech Funding: Smart Lenses and Cross-Border Data The era of speculative consumer health apps has given way to heavily vetted, deep-tech clinical innovations. Azalea Vision Bags EIC Accelerator Funding: The Belgian healthtech firm was selected for the EU's highly competitive European Innovation Council (EIC) Accelerator program. They secured up to €7.5 million (including a €2.5 million grant and a planned €5 million equity investment) to move their medical-grade smart contact lens into clinical trials. The lens treats complex vision issues (like irregular corneas) and functions as a non-invasive biosensing platform to track biomarkers in tears. EU Deploying Millions for Interoperability: A major shift from "pilot programs" to "full-scale cross-border deployment" was highlighted as the European Innovation Council announced the first three winners of its health data interoperability initiative, securing a combined €3.78 million. The projects (including CARDIO-HUB for elderly remote heart monitoring and NEODATA+ for neonatal intensive care data) aim to break down fragmented regional silos. 2. Regulatory Dynamics: Balancing the AI Act & MDR Navigating Europe's complex regulatory framework remains the defining challenge for startups, leading to a massive industry-wide push for simplification. The Overlap Friction: HealthTech developers are grappling with the dual compliance demands of the newly active EU AI Act and the stringent Medical Device Regulations (MDR/IVDR). Industry bodies are aggressively lobbying the European Commission to streamline overlapping rules, which the EU Parliament projects could save the ecosystem up to €3.3 billion annually in administrative bloat. The UK's "International Reliance" Play: Capitalising on mainland Europe's regulatory friction, the UK's MHRA has introduced its draft Medical Devices (Amendment) Regulations. This creates an "International Reliance" pathway, allowing medical device manufacturers with approvals from specific trusted global regulators to fast-track their entrance into the UK market. 3. Commercial Shifts: Workflow Automation Over "Wellness" According to data from the recently published Philips Future Health Index, roughly 65% of European clinicians have actively ramped up their use of AI medical tech to claw back time. Consequently, venture capital is aggressively backing operational and "plumbing" software that directly tackles administrative burnout rather than patient-facing wellness apps. Feature Area Dominant Trend This Week Investor Focus Administrative AI Reducing medical "no-shows," note-taking, and billing automation. High interest (e.g., recent multi-million rounds for clinical workflow tools like TurnUp and OurMind). Surgical & Hardware AI Advanced computer vision for operating rooms and high-tier medical hardware. High valuation premiums for companies with valid MDR certificates. Consumer Wellness General lifestyle tracking and un-regulated wellness apps. Experiencing a funding drought; strict "flight to quality" by VCs. The Key Takeaway: In Europe's current market, technical brilliance isn't enough. The winners right now are startups that can prove a clean data infrastructure compliant with the upcoming European Health Data Space (EHDS) regulations and show clear, measurable ROI to financially strained hospital networks. >>>> The European MedTech landscape this week is defined by concrete regulatory overhauls aimed at cutting administrative red tape, major institutional shifts in managing software, and substantial growth rounds targeted at systemic health infrastructure. 1. Regulatory Overhauls: Clearing the MDR Backlog Following years of friction surrounding the EU Medical Device Regulation (MDR), the European Commission and member states have pushed forward major administrative changes to alleviate structural bottlenecks. The "Digital Omnibus" Sets Clear AI Deadlines: Following the provisional political agreement on the Digital Omnibus (amending the landmark EU AI Act), it has been formally clarified how AI-enabled medical devices are categorized. While standalone software and hardware devices are confirmed as "high-risk," the implementation timeline for high-risk systems under Annex III has been officially deferred from August 2026 to December 2nd, 2027.This delay gives hardware manufacturers a major breathing room. Narrowed "Safety Component" Scope: Crucially, the Digital Omnibus narrows down the definition of high-risk AI. If an AI system embedded in a medical device is used solely for performance optimization, service efficiency, automation, or general quality control, it will no longer be classified as high-risk. "Well-Established Technologies" Simplification: The European Commission published new Delegated Acts for Class III, implantable, and Class IIb devices under its Administrative Simplification Agenda. These updates offer a streamlined clinical evaluation pathway for well-established, legacy technologies, cutting out duplicate clinical trial requirements. 2. Institutional Friction & Global Competitiveness As mainland Europe attempts to patch the operational flaws of the MDR, neighbouring regions are aggressively moving to capture displaced innovation. The Notified Body Fee Clash: As part of the planned revisions to support smaller innovators, the Commission proposed up to a 50% fee discount for micro and small enterprises applying for MDR assessments. However, Team-NB (the association of European Notified Bodies) fiercely pushed back this week, warning that forcing them to subsidise smaller companies will cause a major operational imbalance, increase bottlenecks, and destabilise the assessment system. They are lobbying for structured, free pre-application dialogues instead. 3. Commercial Dynamics & Funding Highs The overarching investment trend in European MedTech has shifted sharply away from early-stage, pre-revenue ideas toward highly scalable "care orchestration" and deeply validated clinical platforms. According to tech ecosystem data, capital is concentrating in regional powerhouses like the UK (€2.5B), Switzerland (~€1.0B), and Finland (€881M). Company / Project Funding Clinical / Operational Focus Semble (UK) £30 Million(Series C) Scaling an open, interoperable clinical platform connecting diagnostics, billing, and lab systems across the UK and France. EU DeepTech Therapeutics €11.57 Million Direct EU grants targeting companies with high Technology Readiness Levels (TRL 6–8) specializing in advanced clinical devices and data reuse. Thena Capital(UK) £45 Million(Debut Fund) A new healthcare-focused fund explicitly backing early-stage MedTech and clinical pathway automation. Nanordica Medical(Estonia) €1.6 Million Fast-tracking clinical rollouts for an advanced, antibiotic-free chronic wound care system. The Structural Shift: The common thread linking this week's data is interoperability. European hospitals are entirely burnt out by single-use, closed-loop medical devices. Investors are aggressively favouring platforms that integrate seamlessly with existing hospital infrastructure while maintaining ironclad compliance with the incoming European Health Data Space (EHDS) frameworks. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

Halfway through 2026, Nelson Advisors predictions on what’s to come in European HealthTech and MedTech Executive Summary The first half of 2026 confirmed the thesis we set out last December: European HealthTech and MedTech are transitioning from a volume-driven market into a value-driven one. Fewer companies are being funded, but the survivors are being funded harder; fewer deals are being signed, but the ones that close are larger, more strategic and increasingly organised around a single question, who owns defensible, clinically validated artificial intelligence. We enter the second half of the year with a market that is quieter on the surface and structurally more active underneath. Our headline call for H2 2026 is that the “bigger cheques, fewer bets” dynamic accelerates. We expect European digital health venture funding to remain well below its 2021–2022 peak in deal count, while average round sizes continue to climb and late-stage capital concentrates in a narrow band of category leaders. On the M&A side, we anticipate a busier second half than first, driven by strategic carve-outs, private-equity buy-and-build platforms, and incumbents using their balance sheets and compliance infrastructure to acquire innovation they can no longer afford to build slowly. The regulatory picture, meanwhile, has shifted in a way that materially changes deal timing, and we think most market participants have not yet fully priced it in. This report sets out ten predictions across capital markets, M&A, artificial intelligence, regulation, sub-sectors, private equity and geography, followed by the principal risks to our view and the strategic actions we believe founders, acquirers and sponsors should take before year-end. The H1 2026 Backdrop Any credible forecast has to start from where the market actually is, and the numbers from the first quarter frame the picture cleanly. European digital health venture funding reached roughly $1.2 Billion across 67 deals in Q1 2026, a decline of around 44% in capital deployed and 46% in deal count against the same period in 2025. On its own that reads as a market in retreat. But the average deal size rose to approximately $21 million, up 8% year-on-year and three mega-rounds of $100 Million or more closed in the quarter, led by Oviva’s $235 Million Series D, Alan’s $116 Million Series G and DentalMonitoring’s $100 Million Series D. This is not a market that has run out of capital; it is a market that has become far more selective about where capital goes. The exit environment tells a complementary story. Thirteen European exit transactions in Q1 2026 carried roughly $552 million in disclosed value, led by Kaia Health at $285 Million and Gleamer at $267 Million. Patient Solutions captured the largest share of new capital at around $298 Million, with Medical Diagnostics close behind and cardiovascular, diabetes and nutrition-focused ventures attracting the deepest therapeutic funding. The signal is consistent: money is flowing to categories with clear reimbursement pathways, demonstrable clinical outcomes and a credible route to either scaled commercialisation or strategic exit. The broader market context remains genuinely large. Europe’s digital health market generated an estimated $130 Billion in revenue in 2025 and is projected to compound at roughly 10% annually toward $314 Billion by 2034, while the European HealthTech market specifically is forecast to grow at an 18% CAGR from around $97 Billion in 2025 toward $222 Billion by 2030. Underneath a soft funding headline sits a structurally expanding end-market. That gap between subdued private financing and robust underlying demand is precisely the condition under which strategic and sponsor acquirers move and it is the foundation for most of what follows. Prediction 1 — M&A gets busier in the second half and the deals get bigger We expect European HealthTech and MedTech M&A activity to be materially higher in H2 2026 than in H1, continuing the shift from cautious, volume-driven dealmaking to high-value, transformative transactions. European healthcare M&A already demonstrated its resilience through 2025, with deal value spiking roughly 87% to €31.8 Billion in the first half of that year even as deal count fell around 8%. That “fewer, larger” pattern is now the base case rather than an anomaly. The catalysts are aligned for a strong close to the year. Large listed healthcare businesses are under pressure to prune their portfolios, and we expect a wave of carve-out and divestiture activity as incumbents shed non-core assets to fund AI and high-growth therapeutic bets. Roughly half of European dealmakers surveyed expect activity to rise over the coming twelve months despite persistent volatility and the macro backdrop, with global M&A projected by some houses to approach record territory in 2026, supports risk appetite at the top of the market. We anticipate a rising number of mega-deals at the $5 Billion-plus level, concentrated in advanced diagnostics, neurovascular & neuromodulation and AI data platforms, alongside a deeper stream of bolt-on transactions in the €25–250 Million range where most of our clients operate. Prediction 2 — Funding stays concentrated, and the mega-round returns selectively On the venture side, we do not expect a broad-based recovery in deal count during H2 2026. Instead, we expect the concentration to intensify: a widening gap between a small cohort of category leaders raising large late-stage rounds and a long tail of early-stage companies facing a genuinely difficult financing environment. Average round sizes should continue to rise even if aggregate capital deployed stays soft, because investors are consolidating conviction into fewer names. The clearest opportunity and risk sits at Series B and the growth stage. Companies with proven unit economics, real reimbursement traction and defensible technology will find capital available on reasonable terms; those still searching for product-market fit or dependent on pilot revenue will face down rounds, bridge financings and in many cases, acqui-hire outcomes. We expect the mega-round (≥ $100 Million) to remain a feature rather than the norm, clustered in obesity and metabolic care, AI-enabled diagnostics and infrastructure and insurance or payer-adjacent platforms. For founders, the practical implication is that raising in H2 2026 will reward demonstrable outcomes data over narrative more than at any point in the last five years. Prediction 3 — AI moves decisively from investment thesis to competitive filter The single most important structural force in both sectors remains the race to acquire defensible artificial intelligence. AI already captured the majority of Europe’s digital health funding through 2024 and that dominance has, if anything, deepened. What changes in H2 2026 is the nature of the advantage: AI stops being a differentiator that attracts a premium and starts being a threshold requirement without which assets struggle to attract capital or acquirers at all. We expect the valuation spread to persist and widen. Companies with proprietary, clinically-validated algorithms and deep integration into clinical workflows should continue to command premium multiples in the region of 6x to 8x revenue, against a broader HealthTech range closer to 4x to 6x. The most sought-after targets remain ambient clinical intelligence and AI scribes, AI-powered diagnostics and medical imaging and revenue-cycle and operational automation platforms. A second-order effect is what we would call the AI deflationary wave: as generative tools compress the cost of building certain software categories, differentiation migrates decisively toward proprietary data assets, regulatory clearances and distribution, the things that cannot be replicated by a model. We expect acquirers to pay up for those moats and to discount undifferentiated software aggressively. Prediction 4 — The regulatory clock resets, changing deal timing more than deal logic The regulatory story is where we think the consensus view is most out of date. The market spent two years pricing in August 2026 as the hard deadline for full high-risk obligations under the EU AI Act, the point at which AI-enabled medical devices would need to complete conformity assessment against a demanding set of data-governance, human-oversight and transparency requirements. That deadline is now moving. Under the European Commission’s proposed “Digital Omnibus” package, with support signalled from both Council and Parliament, the enforcement dates for high-risk systems are expected to shift to December 2027 for standalone systems and August 2028 for AI embedded in regulated products, including medical devices. We read this as a reprieve, not a reversal. The direction of travel, toward stringent oversight of clinical AI, with the parallel application of the AI Act and MDR/IVDR adding an estimated 18 to 24 months to certification timelines for higher-risk software, is unchanged. What changes is timing and breathing room. In the near term this relieves some pressure on notified-body bottlenecks and gives under-resourced companies more runway to achieve compliance. Over the medium term, however, the fundamental competitive dynamic holds: high fixed compliance costs continue to strain under-capitalised SMEs and continue to advantage large incumbents such as Medtronic and Philips that can absorb them. Layered on top sits the European Health Data Space, now in force and rolling out through the decade, which we expect to become a genuine M&A catalyst by creating value around consent-management infrastructure, health-data access intermediaries and AI platforms trained on structured, cross-border data. Our net conclusion is that regulation remains the market’s most powerful consolidation engine and the Omnibus delay simply changes when, not whether, that consolidation plays out. Prediction 5 — Sub Sector calls: where we expect capital and deals to cluster Obesity and metabolic care remain, in our view, the defining investment theme of the year. GLP-1 therapies have reshaped the entire adjacent ecosystem and companies building monitoring, adherence, titration and outcomes platforms around them are becoming strategically valuable to both pharmaceutical manufacturers and payers. Obesity-focused ventures attracted around $300 Million and diabetes a similar figure in Q1 2026 alone and we expect pharmacotherapy plus-digital hybrid models to attract continued capital and acquisition interest through H2 as the market shifts from consolidation toward scaled delivery. Mental health is the second theme we would overweight. Demand continues to outstrip capacity across Spain, the UK and Germany, with waiting times measured in months and the category attracted more capital than any other therapeutic area in the US at the start of 2026. We expect European mental-health platforms with clinical validation and payer contracts to be prime bolt-on targets. In diagnostics and imaging, AI-powered tools remain the most mature emerging technology, with adoption on a steep trajectory; we expect this to be among the most active M&A subsectors of the second half. Ambient clinical intelligence, AI scribes and voice technologies embedded in the clinical encounter, should see accelerating European adoption despite and partly because of, the tighter regulatory frame. Beyond these, we continue to like femtech, preventive and behaviour change care, and the earlier stage but strategically important frontiers of bioelectronic and neuromodulation medicine, sleep technology and defence adjacent MedTech and supply chain resilience, each of which we expect to feature in selective dealmaking. Halfway through 2026, Nelson Advisors predictions on what’s to come in European HealthTech and MedTech Prediction 6 — Private equity becomes the dominant architect of consolidation If strategics set the tone at the top of the market, we expect private equity to do the heavy lifting across the middle. Sponsor buyout activity in European healthcare rose sharply through 2025, up roughly 276% to €29.6 Billion year-to-date against 2024 and PE deal volume reached a record, surpassing the previous 2021 peak. We expect that momentum to carry through H2 2026, with three strategies dominating. The first and most important is buy and build. The structural fragmentation of European HealthTech and MedTech makes it close to an ideal environment for platform consolidation and we expect sponsors to assemble scale through a lead platform acquisition followed by a programmatic series of bolt-ons and technology integration. The second is the AI-native merger, in which PE firms bolt AI-native capabilities onto legacy healthcare businesses to create modern, data-driven platforms, a pattern already visible in revenue-cycle management and one we expect to spread into diagnostics, clinical documentation and care operations. The third is the club deal, where sponsors partner with corporate buyers to concentrate on a specific therapeutic area, sharing risk and combining sector expertise. With substantial dry powder still committed to the sector, including large dedicated early-stage funds, we expect PE to be the most consistent source of liquidity for founders through year end, particularly for Series B and later companies with proven unit economics positioned for an M&A-centric exit. Prediction 7 — Geography: the UK leads, DACH and the Nordics anchor, France stays selective We expect the United Kingdom to retain its position as Europe’s most active single digital health market and to register the strongest structural growth in Health IT. The NHS 10 Year Health Plan’s move toward standardised, value-based procurement, shifting roughly £10 Billion of annual MedTech spend from cost-driven to outcome-driven purchasing, is, in our view, one of the most consequential demand side developments in Europe and it should reward companies that can evidence outcomes rather than merely price. We expect this to draw both domestic and international acquirer interest in UK assets through H2. The Nordics should continue to punch above their weight in AI-driven oncology, preventive health and clinical grade diagnostics, supported by strong data infrastructure and a receptive clinical culture. Germany remains the largest and most consistent Continental market despite some softness in deal value and we expect DACH to be a focal point for both carve-outs and buy-and-build platforms. France and the Netherlands remain steady, selective performers. Across all of these, we expect Europe to keep absorbing relative share from a US market that has seen its own dislocations, though we would caution that US mega-round dynamics continue to set the global valuation benchmark against which European assets are measured. Prediction 8 — Valuation discipline holds, with a widening quality premium We do not expect a broad re-rating of the sector in H2 2026. Instead, we expect the bifurcation in valuation to sharpen. Assets with clinical validation, regulatory clearances, real reimbursement and defensible data or AI will continue to clear at premium multiples and in competitive processes, above ask. Undifferentiated software, pilot-stage revenue and “AI-enabled” positioning without proprietary substance will continue to face compression, longer processes and structured outcomes. For sellers, the gap between a well-prepared, evidence backed process and an opportunistic one has rarely been wider and that gap is, in practice, worth multiple turns of revenue. Prediction 9 — Cross-border and pharma-adjacent capital deepens The pharmaceutical patent cliff, with a very large tranche of branded sales exposed to loss of exclusivity between 2026 and 2030, continues to push Pharma toward innovation-led M&A and toward digital and data assets that extend the value of their franchises. We expect this to manifest in HealthTech through deeper partnerships and acquisitions around companion digital tools, real-world-evidence platforms and adherence infrastructure, particularly in metabolic, oncology and cardiovascular care. Pharma-adjacent capital, in our view, becomes one of the more reliable sources of both partnership revenue and eventual exit for European digital health companies operating close to the therapeutic frontier. Prediction 10 — Preparation, not timing, determines outcomes Our final prediction is less about the market and more about how participants should meet it. In a value-driven, selectivity-first environment, outcomes are determined well before a process begins. The companies that will command premium valuations and attract competitive processes in H2 2026 are those that have already invested in clinical validation, regulatory readiness under MDR/IVDR and the AI Act, interoperability with EHR systems and EHDS standards such as FHIR and OMOP and clean, defensible data assets. We expect the reward for that preparation to be unusually large this year, and the penalty for its absence to be unusually severe. Risks to Our View Several factors could move the market against this base case. A sharper-than-expected macroeconomic deterioration, a spike in rates or a broad risk-off episode would compress both financing and M&A appetite quickly, given how sentiment-driven the top of the market has become. A reversal or further delay in the Digital Omnibus package could reintroduce the compliance-cliff pressure we now expect to ease, changing deal timing. US mega round dynamics could pull capital and talent back across the Atlantic and widen the valuation gap against European assets. And a high-profile failure of a clinical AI product, whether a safety event or a reimbursement withdrawal, could chill enthusiasm for the AI-anchored theses that underpin much of our outlook. We regard each of these as plausible rather than probable, but they define the distribution around our central case. Strategic Implications For founders and sellers, the message is to prioritise evidence over narrative: invest in clinical validation and regulatory readiness now, build interoperability into the product, and prepare processes that put outcomes data at the centre. Category leaders should consider whether H2 2026 is the moment to raise a decisive late-stage round or to run a sale process into strong strategic and sponsor demand. For strategic acquirers, we would focus the second half on regulatory-infrastructure and data plays ahead of EHDS milestones, on AI-native platforms that can be integrated across an existing portfolio, and on disciplined bolt-ons in mental health, femtech, metabolic care and preventive health. For private equity, the environment favours buy and build platforms in fragmented verticals, AI-native mergers that modernise legacy assets, and club deals that concentrate expertise and share risk in attractive therapeutic areas. The through-line across all three audiences is the same one we opened with: capital and conviction are concentrating, AI has become the organising principle of value and regulation is quietly rewiring who can compete. The second half of 2026 will reward those who are prepared, evidenced and clear-eyed about where durable advantage actually sits. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

The Paradigm Shift in European Healthcare M&A Advisory The European healthcare technology (HealthTech) and medical technology (MedTech) advisory ecosystems are undergoing a definitive structural realignment. Driven by the transition from the liquidity-fuelled, growth-at-all-costs environment of the early 2020s to a disciplined, metrics-centric climate, this transformation is characterised as the "Great Rationalisation". In this environment, enterprise valuation is no longer determined by raw revenue expansion, but by clinical utility, regulatory resilience, and seamless integration into established clinical pathways. Consequently, traditional bulge-bracket investment banking institutions are ceding the high-growth mid-market to a sophisticated tier of specialist boutique advisors. These specialist firms are led by "Founder Bankers" and seasoned clinicians who offer direct operational empathy and deep scientific literacy, allowing them to bridge the linguistic and valuation gaps between agile technology founders and risk-averse institutional buyers. This selective recovery is marked by a divergence between transaction volume and upfront transaction value. Strategic acquirers are executing fewer but much larger, high-value platform acquisitions to prioritize proven technology and category leadership over speculative growth. To demonstrate this structural trend, the following data points illustrate the macro capital movements and transaction parameters defining the contemporary European landscape: Metric 2024 Actual 2025 Estimated / Observed 2026 Projected Strategic Significance Global Healthcare M&A Volume $417.8 Billion $450.0 Billion+ $3.9 Trillion (Global All Sectors) Focuses capital allocation on scaled digital platforms and de-risked strategic assets. European Healthcare PE Value $59.9 Billion $80.9 Billion $95.0 Billion+ Rebounds strongly to deploy massive financial sponsor dry powder via buy-and-build consolidation. MedTech Deal Count 41 42 50+ Reflects a stabilized deal volume concentrated in high-complexity clinical platforms. Average MedTech Deal Size $1.6 Billion $795.1 Million (Adjusted) $900.0 Million+ Underscores the consolidation of capital into premium, clinically validated platforms. Median MedTech Upfront Payment $14.0 Million (Q4) $250.0 Million (Q1) To Be Determined [cite: 1, 6] Demonstrates an exponential rise in upfront valuation for de-risked clinical technology. Average HealthTech Deal Size $13.6 Million (Q1 2022) Transition Period $46.6 Million (Q1 2026) Shifts capital from early-stage testing to late-stage platform scale and integration. European Digital Health Funding ~$1.1 Billion (Q1) ~$2.0 Billion (Q1) Post-Recovery Phase Reflects an 82% year-over-year rebound focusing on platform scale and regional integration. Global Digital Health Exits Transition Period 113 Exits (H1 2025) Observation Phase Illustrates the dominance of M&A (107 M&A vs. 6 IPOs, or 94.7%) over public listings. Market Activity, Capital Formation and Private Equity Integration The current cycle is characterised by a "flight to quality," where capital efficiency and proven unit economics are the primary determinants of value. Following the post-pandemic valuation corrections of 2023, the market has settled into a bifurcated state. Premium assets, featuring proprietary clinical AI, robust clinical validation, and clear regulatory certification, command historically high multiples, while secondary assets face severe compression or are forced into defensive consolidation. This bifurcation is further illuminated by the valuation multiples across specific digital health and MedTech asset classes: Asset Class Valuation Metric Range / Multiple Strategic / Valuation Driver Premium AI & Data EV / Revenue 6.0x – 8.0x+ Driven by proprietary algorithms, clean training data sets, and mission-critical clinical workflows. Value-Based Care Tech EV / Revenue 5.5x – 7.0x Anchored in platforms enabling risk-bearing models, cost reduction, and care coordination. Hybrid Telehealth EV / Revenue 5.0x – 7.0x Driven by mature platforms combining virtual triage with physical, in-person clinical capabilities. Standard HealthTech SaaS EV / Revenue 4.0x – 6.0x Reflects growing digital health software with average retention, unit economics, and margins. Profitable HealthTech EV / EBITDA 10.0x – 14.0x Applied to established firms with >20% EBITDA margins conforming to the Rule of 40. Unprofitable / Early-Stage EV / Revenue 3.0x – 4.0x Applied to startups with high burn rates or unclear clinical ROI; experiences severe compression. Private equity has emerged as the primary catalyst for consolidation within the European HealthTech sector. Sponsors leverage "buy-and-build" strategies to consolidate fragmented regional point solutions into unified, pan-European digital platforms. This strategy is illustrated by transactions like Bain Capital's acquisition of HealthEdge, Madison Dearborn Partners' buyout of NextGen Healthcare, and sum-of-assets social care software provider myneva's acquisition by Summa Equity. At the same time, venture capital funding has experienced a stark polarisation. "Mega-deals" exceeding $100 Million account for nearly half of the capital deployed, emphasising the institutional preference for de-risked market leaders with proven clinical traction. Structural Tracks: Industrial MedTech vs. Digital Health SaaS Strategic advisory in the European landscape has bifurcated into two primary, non-overlapping operational tracks: The Industrial MedTech Track The Industrial MedTech track is rooted in physical hardware, clinical robotics, diagnostics, complex imaging, and active implantables. This track is characterized by capital-intensive R&D, extended clinical trial timelines, and exits to large strategic conglomerates like Stryker, Boston Scientific, and Abbott Laboratories. Advisors in this track must possess deep clinical understanding and the capacity to navigate complex regulatory environments, such as the European Union's Medical Device Regulation (MDR/IVDR) and the US Food and Drug Administration (FDA) approval pathways. Value in this track is driven by patent estates, manufacturing scalability, and established reimbursement codes. The Digital Health Track Conversely, the Digital Health track operates on pure technology frameworks, enterprise software scalability, and data monetisation. This segment includes healthcare IT, SaaS-driven clinical software, telehealth and AI-driven diagnostics. Instead of clinical milestones, the performance and valuation of these assets are judged against standardized SaaS software metrics: Net Dollar Retention (NDR), Customer Acquisition Cost (CAC) efficiency, Customer Lifetime Value (LTV) ratios, and recurring revenue churn. The "Rule of 40", where the sum of a company's year-over-year revenue growth rate and EBITDA margin must exceed 40%, is the standard metric for securing premium valuation multiples. Advisors in this track apply technology-first frameworks to help founders bridge the gap between clinical utility and enterprise software scalability. Taxonomy and Profiles of Active European Boutiques and Mid-Market Specialists To navigate this highly fragmented and clinically complex market, a distinct group of boutique investment banks has carved out highly defensible advisory positions. The following table profiles the leading specialist boutiques, tech-focused powerhouses, and mid-market global connectors active across the European HealthTech, MedTech, and Healthcare AI sectors: Specialist Firm Regional Focus & Reach Core Sub-sector Target Focus Key Leadership Executives Highlighted Transactions & Strategic Mandates Nelson Advisors London (HQ), Western Europe, UK, North America. Healthcare AI, Healthcare/Medical Device Cybersecurity, Digital Health, Health IT and FemTech. Lloyd Price (Co-Founder & Partner), Paul Hemings (Co-Founder & Partner). Sourced UK acquisitions for clinical scale-up Evondos; advised Wellola on its strategic sale to a PE portfolio firm. WG Partners London (HQ), pan-European, US market connectivity. Biotech, deep MedTech, active clinical hardware, and Life Sciences tools. Nigel Barnes (Partner), David Wilson (Partner), Claes Spång (Partner). Advised Mereo BioPharma on its $119 million launch and Novartis asset acquisition; Rezolute ($96.9m); Imricor (A$70m raise); Scancell (£11.3m). Clipperton Paris (HQ), Berlin, Munich, New York. Digital Health, Healthcare IT, Clinical SaaS, Telehealth, and HR clinical platforms. Nicolas von Bülow (Managing Partner), Antoine Ganancia (Managing Partner). Advised Hublo on its strategic investment from Five Arrows (Rothschild & Co’s private equity fund). Arma Partners London (HQ), Munich, APAC reach via Latimer Partners. Digital Health, Healthcare IT, B2B Social Care software, and Workforce Management. Paul-Noël Guély (Founder & Managing Partner), Daniel Fugmann (Partner). Appointed by CVC to facilitate the projected £920m sale of System C Healthcare; sum-of-assets social care software provider myneva's acquisition by Summa Equity. Van Lanschot Kempen Amsterdam / Benelux, pan-European, US capital markets. Biotech, MedTech, Diagnostics, Social Care B2B SaaS, and InsurTech. Jan De Kerpel (MD), Nadine Maalouf (MD), Robert-Jan van der Vorm (Director). Acted as Joint Bookrunner on Valneva SE’s €84m reserved offering; advised Khonraad on its 100% sale to Visma; dacadoo Series B and C rounds. TH Healthcare & Life Sciences Global reach across 9 countries, Europe, North America, APAC. Digital Health, HealthTech, mHealth, Wearables, and MedTech product consulting. Vivek Subramanyam (Founder). Advised Aqurance S.A. on its strategic sale to EY; Design + Industry on its sale to Capgemini; C-Clear Partners & Atom Ideas sale to Valantic. DAI Magister London, emerging markets, Africa, Middle East, MENA. Cross-border HealthTech, data/AI-driven healthcare, and remote patient monitoring. Steve Bachmann (Head of US & Co-Head of Europe). Acted as financial advisor to Enterprise AI specialist Fusemachines Inc. on its $200 million Nasdaq listing; advised Satel Oy on its sale to Topcon. Artis Partners London, Munich, New York. B2B AI, deeptech, health AI, FemTech, and D2C telehealth. Founded by the partners of Arma Partners and DAI Magister. Advised Mindler on its acquisition of ieso Digital Health UK; advised Zaptic on its acquisition by Intellect. ConAlliance Munich, DACH region, pan-European. Healthcare-only M&A, European MedTech, clinical services, and digital assets. Executive Managing Partners. Mandated on mid-market DACH MedTech exits and clinical platform sales. Carlsquare Munich, Frankfurt, Stockholm, London, San Francisco. Medical Devices, Implants, Biotech, Diagnostics, and Digital Health. Caspar Graf Stauffenberg (Managing Partner), Anders Bo (Managing Partner). Advised Dr. Willmar Schwabe on its strategic investment in Synaptikon (NeuroNation MED DiGA); advised DENA A/S on the sale of BiopSafe to MedCap AB. Hampleton Partners London, Frankfurt, Stockholm, San Francisco, Shanghai. Healthcare Vertical Software, Health IT Services, EHR, and Medical Hardware. Tom Schmähling (MD & Practice Head), Jonathan Simnett (Director). Advised the shareholders of Berlin-based BaseCase on its sale to EQT portfolio company Certara. Oppenheimer Pan-European reach, US-transatlantic connectivity. Health Tech (EMR, RCM, Telemedicine), MedTech, and Diagnostics. Head of Healthcare Investment Banking. Acted as Exclusive Financial Advisor to Nurami Medical Ltd. and advised Vandemoortele on its acquisition of Lizzi srl. Cain Brothers KeyBanc division, US-transatlantic reach. Payers, providers, services, health IT, and life sciences. Head of Healthcare M&A. Facilitated over 200 healthcare M&A transactions with $45 billion+ in transaction value since 2019. Bishopsgate Corporate Finance United Kingdom / Europe. Small-to-mid market healthcare services, clinical providers, and Health IT. Managing Partners. Delivered lower-mid market exits and strategic corporate carve-outs over 27 years of operations. Macro Transactions, Sector Convergence and Key Valuation Catalysts The integration of technology into clinical settings has catalysed a series of landmark mid-market transactions that define the contemporary consolidation wave. This convergence of clinical delivery and software is illustrated by several high-profile deals: The scale of these transactions is exemplified by the proposed sale of System C Healthcare. Under CVC Capital Partners’ ownership, the asset integrated AI capabilities through the acquisition of FormFlow AI to automate clinical documentation, alongside acquiring Australian community care specialist MYP Technologies to diversify its payer base. As Arma Partners prepares the asset for sale in 2026 on a projected EBITDA of £46 million, the transaction represents a key valuation benchmark for primary data assets. This is because platforms controlling localized, clinical-grade documentation hold the essential datasets required to train clinical AI engines, commanding significant valuation premiums. Similarly, in the deep MedTech and surgical materials segment, New York-listed H.B. Fuller's £715 million acquisition of London-listed Advanced Medical Solutions (AMS) illustrates the strategic premium placed on high regulatory-based barriers to entry and durable clinical demand. The acquisition price of 285 pence per share represents an EV/EBITDA multiple of 12.9x based on consensus forecasts for 2026. This transaction extends H.B. Fuller's portfolio across tissue-bonding adhesives and formulated biosurgicals, while granting it access to AMS's pan-European clinical salesforce and distribution networks. This highlights the trend of industrial conglomerates utilising acquisitions to bypass long organic development and regulatory cycles. Further transaction momentum is illustrated by Siemens’ $5 billion acquisition of life sciences software developer Dotmatics, which represents the largest HealthTech transaction in recent periods. This deal, alongside the strategic investments by private equity firms such as Bain Capital (acquiring HealthEdge for payer infrastructure) and Madison Dearborn Partners (buying out NextGen Healthcare), emphasises the premium placed on scaled digital platforms that control clinical workflows and data pipelines. These consolidation plays are mirrored in the venture capital sphere, where large-scale growth rounds are concentrated in clinically validated, scalable platforms, such as Transcarent’s $700 million and Cleerly’s $500 million raises. This convergence is also reshaping the clinical software and consumer-facing health sectors: NeuroNation MED (Synaptikon): Advising the pharmaceutical giant Dr. Willmar Schwabe Group on its strategic investment in Synaptikon, Carlsquare successfully positioned the permanently approved Digital Health Application (DiGA) NeuroNation MED as a scalable cognitive therapy platform. The transaction represents a structural shift toward combining traditional pharmacology with personalised, reimbursable digital therapeutics. BiopSafe: By advising the Danish family office DENA A/S on the sale of BiopSafe to Sweden’s listed MedCap AB, Carlsquare navigated a niche MedTech device transaction. BiopSafe’s patented closed, formalin-free biopsy handling container directly addresses occupational health challenges in clinical pathology, illustrating how simple hardware innovations with robust safety moats attract public-market acquirers. BaseCase: Highlighting Hampleton Partners’ capabilities in life sciences SaaS, the firm advised the shareholders of Berlin-based BaseCase on its sale to EQT-backed Certara. BaseCase’s interactive data visualisation platform allows pharmaceutical and MedTech companies to effectively communicate the health-economic value of their clinical pipelines to hospital C-suites and payers, accelerating drug commercialisation. The Paradigm Shift in European Healthcare M&A Advisory De-Risking the Clinical-AI Spectrum and Cybersecurity Moats The integration of artificial intelligence into clinical diagnostics, workflow automation, and predictive analytics has redefined the technological moats required to protect enterprise value. However, corporate buyers and private equity sponsors have moved past speculative growth narratives to conduct intensive due diligence on algorithm defensibility and training methodology. Under this analytical framework, "wrapper" companies, which merely place a customized user interface over third-party APIs—are heavily discounted. Conversely, platforms featuring proprietary, clinically validated algorithms trained on clean, proprietary longitudinal datasets command a de-risking premium. To illustrate these technical trends and the key regulatory considerations shaping transactions, the following areas represent critical due diligence requirements for contemporary buyers: Regulatory Moats and the EU AI Act Compliance The European regulatory landscape stands at a key inflection point driven by the overlapping compliance requirements of the EU AI Act and the Medical Device Regulation (MDR/IVDR). For software developers and MedTech manufacturers, navigating these dual frameworks has created a significant capital and operational bottleneck. Specialist boutiques, such as Nelson Advisors, utilise these regulatory hurdles as key valuation drivers. They argue that an asset that has fully secured its regulatory clearance under both MDR and the EU AI Act has built an unassailable regulatory moat. Because a fast-following competitor would require years and millions in clinical testing to replicate these regulatory clearances, a fully compliant technology stack commands a de-risking premium from strategic buyers looking to bypass development lag and enter the European single market immediately. Healthcare and Medical Device Cybersecurity as a Transaction Driver Connected clinical networks, electronic health records, and internet-of-medical-things (IoMT) diagnostic devices are prime targets for sophisticated security breaches. Consequently, robust cybersecurity architecture and proactive device compliance have transitioned from basic IT checklists to critical transaction drivers. Buyers conduct detailed technical due diligence on the integrity of software code, encryption protocols, and clinical data pipelines. A platform that fails to demonstrate rigorous compliance with healthcare data protection frameworks, such as GDPR and the European Health Data Space (EHDS), is viewed as a significant liability. Conversely, companies that embed secure-by-design principles into their medical devices and clinical software protect their enterprise valuations from post-transaction regulatory fines, intellectual property theft, and clinical litigation. Strategic Exits, Series B Gapssand Venture-to-Venture Consolidation A key structural challenge facing European HealthTech startups is the widening "Series B gap". While ample early-stage capital and specialised debut funds, such as Sofinnova Partners’ Capital XI and Thena Capital’s debut vehicle—remain active in seed and Series A rounds, the average timeline to close a Series B now approaches 30 months. This funding gap is driven by late-stage venture capital and private equity investors demanding clear evidence of clinical efficacy, established public reimbursement pathways, and a highly predictable path to profitability. To navigate this liquidity squeeze, specialist boutiques have pioneered the trend of "Venture-to-Venture M&A" and strategic horizontal consolidation. Instead of pursuing dilutive down-rounds, early-stage venture-backed point solutions are consolidating to build comprehensive, multi-product technology platforms. This strategy allows consolidated entities to eliminate administrative redundancies, combine customer acquisition budgets and present a unified clinical product that satisfies the investment criteria of late-stage private equity sponsors. Illustrating this trend is the acquisition of primary care digital triage platform eConsult by digital health scale-up Huma. eConsult, which serves over 1,800 general practice clinics and has delivered over 50 million digital consultations, provides a critical entry point for automated patient triage. By integrating eConsult's triage technology into its "Huma Workspace" clinical platform, Huma created an integrated patient pathway that guides individuals from initial digital intake to automated remote patient monitoring and virtual ward environments. This horizontal integration addresses point-solution fatigue for clinical providers and creates a scalable, high-margin platform. A similar consolidation play is illustrated by digital mental health platform Mindler's acquisition of ieso Digital Health UK, advised by Artis Partners. This transaction combined ieso's NHS relationships and clinically proven, data-driven therapy platform with Mindler's broader European footprint. By unifying ieso’s capabilities with Mindler's existing operations, the combined entity successfully created a pan-European leader capable of delivering mental health treatment across the full spectrum of symptom severity. This transaction highlights the strategic role of specialist boutiques in structuring horizontal combinations that consolidate regional clinical footprints into scalable, pan-European digital platforms. Strategic Synthesis and Practical Recommendations The European healthcare technology and medical technology advisory ecosystems have entered a period of definitive industrial maturity. The era of speculative, growth-at-all-costs capital deployment has been replaced by a disciplined, metric-driven environment where strategic asset value is dictated by clinical utility, regulatory resilience, and technological defensibility. For corporate boards, founders, and financial sponsors, navigating this reorganised landscape requires a clear understanding of the following strategic imperatives: Sponsor-Ready Metric Optimisation: Founders and corporate development teams must align their digital health platforms with standardised SaaS metrics, such as Net Dollar Retention (NDR) and Customer Acquisition Cost (CAC) efficiency, long before initiating a transaction process. Achieving compliance with the Rule of 40 is the standard prerequisite for securing premium valuation multiples from technology buyers. Proactive Regulatory Compliance as a Value Driver: In an environment characterized by the overlapping compliance requirements of the EU AI Act and the Medical Device Regulation (MDR/IVDR), securing full clinical and regulatory clearance must be prioritised as a core value driver. A fully compliant clinical-AI or medical device stack acts as an unassailable regulatory moat, allowing sellers to command significant de-risking premiums from international acquirers. Consolidation of Point Solutions: Hospital systems and clinical payers are experiencing significant point-solution fatigue, prompting them to prioritise unified platforms over isolated, single-feature software tools. Financial sponsors and corporate buyers should actively execute horizontal "buy-and-build" roll-up strategies, combining complementary point solutions—such as digital intake, clinical documentation, and remote monitoring—to build scalable, pan-European digital platforms. Specialist Boutique Advisor Selection: The selection of an M&A advisor is no longer a function of prestige alone, but of strategic alignment with specific operational tracks. While large-cap bulge brackets remain essential for executing multi-billion-dollar transformative mergers, the mid-market is the domain of specialist boutiques. Founders and institutional sponsors must partner with boutique advisors who possess deep, internal scientific and clinical literacy to ensure their technology stacks are effectively positioned and de-risked during technical due diligence. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

The Invisible Infrastructure of Healthcare: Mapping the Socio Technical Architecture and Governance Risks of Shadow AI The rapid, unregulated integration of artificial intelligence into the core workflows of modern medicine has established a pervasive and largely invisible infrastructure of unauthorized technology. Driven by systemic clinician burnout and administrative overload, clinical and administrative staff have increasingly bypassed traditional information technology procurement pathways. This self-directed adoption, collectively termed "Shadow AI", now operates at every stratum of the healthcare hierarchy, from the administrative back office to the surgical suite. Unlike the deterministic Shadow IT of the past, Shadow AI represents a fundamental ontological shift toward non-deterministic, probabilistic systems capable of learning, generating novel content and acting with autonomous agency. This shift compromises established parameters of cybersecurity, information governance, patient safety, and regulatory compliance. The Ontological Shift: Differentiating Legacy Shadow IT from Shadow AI To design effective enterprise governance, healthcare leadership must first distinguish the conceptual and technical boundaries that separate legacy Shadow IT from Shadow AI. Historically, Shadow IT referred to the unauthorised adoption of deterministic software, such as using unapproved cloud storage or collaboration tools to bypass operational friction. While these utilities introduced security vulnerabilities regarding data leakage, the software itself behaved predictably; it did not generate new content, alter data structures, or make clinical decisions. Shadow AI represents a fundamental shift. It involves the use of non-deterministic systems—primarily Large Language Models (LLMs) and autonomous machine learning algorithms—capable of generating novel outputs that may not be grounded in reality. Shadow AI functions effectively as "Shadow Staff," executing administrative and cognitive tasks that were once the exclusive domain of trained medical professionals. This transition creates a dynamic and continuous risk profile. Once Protected Health Information (PHI) is uploaded to an unauthorised consumer LLM, the data can be incorporated into the vendor’s public model training pipeline, creating a permanent, irreversible privacy exposure. Technical Dimension Legacy Shadow IT Shadow AI in Healthcare System Behaviour Deterministic; predictable inputs and outputs. Non-deterministic; probabilistic, generative, and stochastic. Primary Identity Human user accounts. Non-human identities (NHIs), service accounts, and OAuth tokens. Primary Mechanism Unsanctioned SaaS subscriptions and personal hardware. Web portals, browser extensions, IDE plugins, and direct API calls. Data Risk Profile Static data at rest in unauthorised storage repositories. Continuous data processing, inference-time leakage, and model training contamination. Bypass Gateway Bypasses local software distribution policies. Bypasses secure web gateways, firewalls, and CASB solutions. Agency and Labour Logistical utility; does not execute cognitive tasks. Cognitive labour; autonomously drafts notes, synthesises records, and suggests diagnoses. The systemic vulnerability of this interconnected clinical infrastructure was illustrated on December 14, 2025, when DXS International, which provides clinical decision support for approximately 10% of all NHS referrals in England, suffered a data breach impacting its office servers. Although front-line services remained operational, the incident highlighted how third-party risks cascade through the NHS Health and Social Care Network (HSCN). In an ecosystem where cyber, privacy, and AI risks have converged, a single administrative breach can expose integrated clinical networks, highlighting the vulnerability of modern healthcare technology. Socio-Technical Catalysts: Clinician Exhaustion and the Enterprise Gap The widespread adoption of Shadow AI in clinical environments is not driven by employee defiance, but by structural deficiencies and clinical survival mechanisms. Frontline medical staff operate in high-pressure environments characterised by acute administrative overload. The Administrative Crisis and "Pajama Time" The implementation of modern Electronic Health Records (EHRs) has significantly increased clinical documentation requirements. Studies demonstrate that for every single hour a physician spends in direct, face-to-face contact with a patient, they must dedicate an additional two hours to charting and EHR data entry. This massive burden forces clinicians to complete administrative tasks during their personal hours—a socio-technical phenomenon documented as "pajama time". Off-the-shelf consumer generative AI tools offer immediate relief, prompting a behavioral shift where speed and workflow efficiency become the primary drivers of technology adoption. The Enterprise Functionality Gap A major chasm exists between the consumer-grade technologies clinicians utilize in their personal lives and the legacy enterprise infrastructure provided by healthcare organizations. While providers routinely interact with sophisticated, conversational, and highly intuitive AI assistants on their personal devices, hospital workstations often run antiquated EHR interfaces. Approximately 24% to 27% of healthcare professionals who utilise unauthorised AI tools do so because the public platforms provide functionality that is superior to the sanctioned tools available within their organisations. The Staffing Vacuum and "Shadow Staffing" The World Health Organization projects a global healthcare workforce shortage of approximately 10 million providers by the year 2030. In response to rising operational costs, healthcare delivery networks have systematically reduced administrative support personnel and medical scribes. Shadow AI step-functions into this vacuum, acting as virtual digital assistants, coding specialists, and diagnostic guides. This is particularly pronounced in resource-constrained environments such as rural clinics and medical deserts, where a solo practitioner may have no other administrative support. The Policy and Communication Disconnect The escalation of Shadow AI is further accelerated by a communication gap between executive leadership and frontline healthcare workers. While 42% of healthcare administrators believe their AI governance policies are clearly communicated, only 30% of clinical providers agree. Furthermore, administrators are three times more likely to be involved in policy development than the clinicians who actually interact with patients and utilise these tools. This division creates an administrative blind spot where leadership assumes a secure environment, while providers bypass legacy IT controls to manage their daily workloads. Shadow AI Metric Systemic Value Healthcare Implication Average Healthcare Breach Cost $10.93 Million. The highest average breach cost of any sector globally. Shadow AI Breach Cost Premium +$670,000. Added cost due to data exfiltration complexity and third-party model contamination. Insider Negligence Risk Cost $10.3 Million annually. Comprises 53% of total enterprise insider risk, driven primarily by unapproved AI use. Clinician Encounter Rate 40% to 57%. Over half of surveyed clinicians have encountered or used unauthorized AI. Active Employee AI Adoption >80%. Over four-fifths of employees utilize unapproved AI platforms. Shadow AI Detection Lag 247 Days. Six days longer than standard data breaches, increasing exposure time. Unapproved Platform Inventory 665 distinct applications. The volume of unapproved generative AI applications tracked across enterprises. Organizational Policy Deficit 63% lack AI governance. Only 37% of organizations have formal policies to manage or detect AI risk. Clinical Workflows and Patient Safety Hazards To bypass traditional hospital IT security boundaries, clinical staff have developed several surreptitious workflows that utilise consumer-grade devices, personal browsers, and direct APIs. These workflows introduce severe clinical and data protection liabilities. The Air-Gapped Tripartite Architecture This workflow operates across three distinct architectural layers designed to bypass enterprise security boundaries: The Input Layer: A clinician dictates a clinical note onto their personal mobile device or manually copies structured patient records from a secure, networked workstation. The Processing Layer: The provider pastes this raw PHI into a browser-based consumer AI portal (such as ChatGPT, Claude, or Gemini) hosted on external, public servers. The Output Layer: The public model processes the data, retains it for future training cycles, and returns a synthesised note. The clinician then copies this output and pastes it directly back into the secure hospital EHR. This workflow is difficult to detect because it is entirely air-gapped from the network security layer of the hospital’s core infrastructure, leaving no audit trails in the EHR logs. The Ambient "Digital Scribe" Workflow Clinicians use personal smartphones or unapproved web applications to record entire, live patient encounters. The audio transcript is processed through commercial speech-to-text engines and subsequently routed through consumer generative models with instructions to compile a structured Subjective, Objective, Assessment, and Plan (SOAP) note. This captures raw acoustic biometrics and intimate diagnostic disclosures, transmitting them to external vendors without patient consent, secure audit trails, or the Business Associate Agreements (BAAs) required for HIPAA compliance. Surreptitious Clinical Decision Support Beyond administrative tasks, clinicians utilise unauthorised models to guide clinical decisions. Clinicians enter complex patient histories, laboratory values, and drug regimens into consumer interfaces to generate differential diagnoses or screen for complex drug-to-drug interactions. Because general-purpose models lack specialized medical knowledge bases, clinical validation layers and safety guardrails, this workflow exposes patients to significant diagnostic errors. Documented Safety and Clinical Diagnostics Failures The clinical efficacy of general generative AI models remains highly volatile, and their use in diagnostic decision support introduces significant risks. The "Bixonimania" Fabricated Pathology: In 2024, Swedish researchers created a fabricated medical condition named "bixonimania" to evaluate model verification protocols. Within weeks, major consumer platforms—including ChatGPT, Google, Copilot, and Perplexity—consistently diagnosed patients with this non-existent disease and generated plausible-sounding pathophysiological mechanisms for it. These AI-generated falsehoods eventually propagated into peer-reviewed clinical publications. Severe Emergency Under-Triage: A clinical evaluation published in Nature Medicine revealed that ChatGPT Health under-triaged 52% of active emergency presentations. The model proved highly susceptible to conversational framing; if a patient's prompt included language indicating that their family or friends were minimizing their symptoms, the algorithm was 11.7 times more likely to recommend lower-acuity, non-urgent care, even in scenarios involving life-threatening cardiac or neurological emergencies. Systemic Diagnostic Failures: An evaluation published in JAMA Network Open tested 21 leading AI models (including GPT-5, Claude, and Gemini) against complex clinical scenarios. The differential diagnosis failure rate exceeded 80% across all evaluated models, demonstrating that general-purpose LLMs are currently unsuitable for unguided clinical decision support. The Patient-Led Ecosystem: Consumer Shadow AI and the Liability Gap A parallel dimension of this issue is the rapid adoption of "Consumer Shadow AI," where patients independently consult public AI assistants to bypass formal clinical pathways. OpenAI reports that out of 800 million monthly active users on ChatGPT, over 230 million people utilize the platform weekly for health and wellness inquiries. Uptake rates for health-related queries range from 9.9% of consumers in Australia to 32.6% in the United States. Mental Healthcare Access and Digital Triage Consumer Shadow AI is heavily utilised as an informal alternative for psychological support. Approximately 48.7% of AI assistant users report consulting these systems to address anxiety and depression, with 63.4% reporting perceived improvement in their mental health. Patients are drawn to these tools due to their convenience, rapid synthesis of complex data, and perceived empathy. In evaluations of Google's Articulate Medical Intelligence Explorer, patient-actors rated the AI assistant higher than human physicians on 25 out of 26 communication dimensions, including rapport building, politeness, and active listening. However, the use of unvetted models can lead to dangerous clinical outcomes, illustrated by Google’s forced removal of medical summaries from its "AI Overviews" feature following public safety failures. The Liability Gap A stark legal disconnect exists between the marketing of AI assistants for healthcare and the contractual realities defined in vendor terms of service. While major AI developers promote healthcare-specific tools, their terms of service explicitly state that the platforms do not provide medical advice and are intended strictly for general informational purposes. This shifts the entire burden of clinical risk to the consumer or the individual physician. Furthermore, standard enterprise contracts limit vendor liability to minimal amounts, creating a significant legal exposure for healthcare organisations. AI Vendor / Platform Typical Health & Wellness Product Terms of Service Disclaimer Maximum Contractual Liability Cap OpenAI ChatGPT Health / GPT-4o General information only; not a substitute for professional medical advice. The greater of 12 months of fees paid or $100. Anthropic Claude for Healthcare General information only; does not establish a clinical relationship. The greater of 6 months of fees paid or $100. Google Med-PaLM 2 / Gemini Information only; does not replace qualified medical decision-making. The greater of fees paid or $500 (or 125% of fees). Technical Surfaces of the Invisible Infrastructure: Agents, WebSockets and OAuth Token Sprawl The technical risk profile of Shadow AI has expanded beyond manual copy-paste workflows into web-based chatbots. The modern threat landscape is defined by an invisible infrastructure consisting of API sprawl, unauthorised browser extensions, and autonomous "Shadow AI Agents" that operate inside trusted network perimeters. The Shift to Agentic Autonomy Traditional generative AI tools are reactive, requiring direct human inputs to generate specific outputs. The risk is limited to the data a human chooses to share in a single interaction. However, Agentic AI introduces autonomous systems that execute multi-step workflows without human intervention. Once deployed, these agents continuously access data, connect with other SaaS platforms, make real-time decisions, and perform transactions at machine speed. Because they are not cataloged in standard enterprise registries, they operate without oversight, logging, or human validation. Browser Extensions and the "Co-Pilot" Attack Vector Lightweight browser extensions with integrated generative AI capabilities represent a primary entry point for unmanaged AI inside clinical networks. Staff install these extensions to summarize clinical journals, draft patient emails, or auto-complete documentation within web-based EHR interfaces. Because these extensions require extensive browser permissions, such as the ability to read and modify all data on visited websites, they can continuously parse EHR screens, capture patient records, and exfiltrate data to unvetted external APIs. Non-Human Identities (NHIs) and OAuth Token Sprawl Many modern AI tools allow users to bypass complex procurement workflows through single-click OAuth integrations with corporate accounts. This creates a network of Non-Human Identities (NHIs). These third-party AI systems are granted broad read-and-write permissions to cloud ecosystems and corporate repositories. These tokens often remain active indefinitely, even after the employee has stopped using the tool or has been offboarded from the organization, creating persistent, unmonitored pathways into sensitive databases. Model Context Protocol (MCP) and CI/CD Pipeline Infiltration Software developers and system architects within healthcare networks introduce shadow risks by integrating Model Context Protocol (MCP) servers into their environments. Developers utilize unvetted AI code assistants and IDE extensions (such as Cursor) that connect directly to production databases and code repositories to compile analytics or generate scripts. This introduces the risk of code supply chain contamination, model-poisoning and the inadvertent exposure of clinical database structures. The Invisible Infrastructure of Healthcare: Mapping the Socio Technical Architecture and Governance Risks of Shadow AI The Regulatory and Compliance Minefield The use of unapproved AI tools introduces severe legal and financial liabilities across global regulatory frameworks. Because these laws assume controlled, documented data processing, the use of unmanaged AI creates compliance challenges. The Health Insurance Portability and Accountability Act (HIPAA) The regulatory anchor of healthcare privacy in the United States is the HIPAA Security and Privacy Rules. The Business Associate Agreement (BAA) Requirement: Any third-party utility that processes, transmits, or stores Protected Health Information must sign a legally binding BAA. Because consumer-grade AI platforms explicitly disclaim clinical liabilities and do not provide BAAs for free or standard consumer accounts, any transmission of PHI into these interfaces is a direct, actionable HIPAA violation. Minimum Necessary Standard: HIPAA requires organizations to limit the exposure of PHI to the absolute minimum necessary to complete a task. Shadow AI tools, which ingest complete clinical notes or entire transcripts, violate this standard. Penalties and Liability: Civil monetary penalties can reach up to $1.5 million per violation category per year, alongside potential criminal charges for willful neglect and professional licensure challenges. The General Data Protection Regulation (GDPR) and UK GDPR Under GDPR, patient health metrics are classified as "Special Category Data," triggering high levels of statutory protection. The Article 28 Data Processing Agreement (DPA): Processing personal data requires a DPA that outlines processor obligations, security controls, and breach notification windows. Consumer AI tools operating under standard terms of service fail to meet these Article 28 requirements. The Article 22 Prohibition on Automated Decisions: GDPR explicitly restricts individuals from being subject to decisions based solely on automated processing that produce legal or similarly significant effects. Shadow AI clinical tools used without human verification violate this principle. Mandatory Data Protection Impact Assessments (DPIA): Under GDPR, processing special category health data on a large scale via algorithmic systems requires a DPIA prior to deployment. Bypassing this step constitutes an independent regulatory violation. The European Union Artificial Intelligence (EU AI Act) Taking effect through a staggered implementation timeline, the EU AI Act establishes a strict risk-based framework for AI applications. High-Risk Classifications (Annex I and Annex III): Most AI systems used in healthcare—including those assisting with diagnostics, triage, patient monitoring, and clinical decision support—are classified as "High-Risk". The August 2026 Mandate: By August 2026, healthcare providers deploying high-risk systems must meet stringent compliance standards, including documented risk management, accuracy testing, bias auditing, cybersecurity baselines, and human oversight controls. Deployer Obligations: Hospitals and clinical networks acting as deployers must ensure human oversight, maintain operational logs, provide AI literacy training to staff, and conduct a Fundamental Rights Impact Assessment (FRIA). Staggered Regulatory Timeline: February 2, 2025: Absolute prohibitions on unacceptable-risk practices (such as manipulative behavioral profiling or untargeted biometric scraping) became enforceable. August 2, 2025: Obligations for General Purpose AI (GPAI) providers (such as model cards, technical documentation, and training data transparency) took effect. August 2, 2026: Comprehensive compliance requirements for high-risk clinical and diagnostic deployments become fully enforceable. December 2, 2026: Transparency obligations for AI-generated content providers apply. December 2, 2027: Compliance obligations extend to standalone high-risk systems listed under Annex III. The California Consumer Privacy Act (CCPA) Under the CCPA, shadow AI tools trigger major compliance violations when patient inputs are used for model training. This process qualifies as selling or sharing personal data without explicit consumer disclosure or opt-out rights. Furthermore, because data ingested into consumer LLMs becomes part of a distributed model, satisfying a consumer's right to delete becomes technically impossible. Regulatory Requirement HIPAA (United States) GDPR (European Union) Contractual Standard Business Associate Agreement (BAA). Data Processing Agreement (DPA) under Article 28. Special Category Treatment Protected Health Information (PHI). Special Category Data under Article 9. Risk Assessment Security Risk Assessment (SRA). Data Protection Impact Assessment (DPIA). Transit Encryption TLS 1.2 or TLS 1.3 mandated. Mandatory state-of-the-art encryption. Rest Encryption AES-256 mandated. Mandatory state-of-the-art encryption. Breach Notification Timeline Within 60 calendar days of discovery. Within 72 hours of becoming aware. Audit Log Retention Mandatory 6-year to 7-year retention. Mandatory processing records under Article 30. Consumer Erasure Support Not applicable (clinical record laws supersede). Mandated under Article 17 (subject to clinical exemptions). A Unified Architectural Framework for Shadow AI Mitigation Managing Shadow AI requires a transition from reactive blocking to proactive governance and the enablement of secure alternatives. Healthcare delivery networks can address this challenge through several coordinated strategies: 1. Multi-Layered Technical Detection Healthcare security teams cannot rely on a single defensive layer to identify unmanaged AI systems. Effective discovery requires a coordinated technical approach: Network Infrastructure Monitoring: Security teams must analyze DNS queries, TLS handshakes, and outbound connection logs to identify traffic routing to known consumer AI domains (such as openai.com, claude.ai, or gemini.google.com) and external API endpoints. Browser-Native point-of-interaction Security: Because shadow AI operates primarily within browser sessions, organizations should deploy lightweight browser-native security tools. These tools intercept user actions—such as copy-pasting text, uploading clinical PDFs, or submitting screenshots—before the data leaves the organization's control. Documentation Pattern Analysis: Security groups can deploy Natural Language Processing (NLP) tools to audit EHR documentation for telltale signs of machine-generated text. These signs include unusually formal or highly consistent phrasing across different clinicians, suspiciously rapid documentation creation, or specific AI phrases (e.g., "As an AI language model..."). SaaS and Identity Auditing: Organizations must scan identity provider logs and cloud environments to inventory unauthorized OAuth tokens, third-party API integrations, and non-human identities operating within production systems. 2. The Integration of DSPM and DLP ("DSPM First, DLP Last") Traditional security tools are designed to secure static data at rest. However, Shadow AI represents data in motion, moving dynamically across clinical networks. Addressing this requires a unified Data Security Posture Management (DSPM) and Data Loss Prevention (DLP) architecture: DSPM Role: The DSPM engine acts as the baseline, continuously discovering, classifying, and mapping patient data across all enterprise storage, databases, and pipelines. It establishes context by identifying where PHI resides and who has permissions to access it. DLP Role: The DLP engine acts as the real-time enforcement mechanism, inspecting outbound data flows against the classification rules defined by the DSPM layer. It monitors exfiltration pathways—such as copy-pasting, uploading files, or running web-based scripts—and intervenes before data exits the clinical environment. Advanced Monitoring Capabilities: To enforce these policies, the integrated system must support WebSocket monitoring to inspect active streaming text, Optical Character Recognition (OCR) to detect PHI within image files or screenshots, and Exact Data Match (EDM) to verify clinical terms against active patient records. 3. Policy and Organisational Governance Technical detection must be paired with structured organizational policies: Acceptable Use Policies: Healthcare networks must establish explicit, practical AI usage policies that define approved tools, prohibited platforms, acceptable use cases, and standard procedures for requesting new tool approvals. Establish Cross-Functional Governance Bodies: Organizations should form dedicated AI governance committees that include clinical leaders, IT security teams, and compliance officers. This committee maintains the approved tool catalog, evaluates new systems, and coordinates employee upskilling and peer mentoring programs to reduce unapproved technology use. AI Compliance Officer: Larger health systems should establish a dedicated AI Compliance Officer to manage conformity assessments, supervise risk management, maintain technical documentation, and act as a liaison with regulators. 4. Transitioning Staff to Sanctioned Solutions The most effective way to eliminate Shadow AI is to provide approved, secure, and compliant solutions that solve clinicians' administrative problems while maintaining strict data controls. These tools must operate under signed Business Associate Agreements, restrict data from being used for public model training, and provide complete, clinical-grade audit logs. Sanctioned AI Solution Target Practice & Clinical Setting EHR Integration Model Primary Pros Core Cons & Implementation Challenges Nuance DAX Copilot (Microsoft) Large health systems, academic medical centers, and Epic users. Deep, native integration with Epic and major enterprise EHRs. Strongest native EHR integration; backed by Microsoft infrastructure. Highly expensive ($500–$1,500/provider/month); complex enterprise rollout. SOAPNoteAI Independent practices, small groups, and multi-specialty clinics. Copy-paste workflow compatible with any standard EHR. Highly affordable; no long-term contracts; supports multiple input methods. Lacks deep native EHR integration, requiring manual copy-pasting. athenaAmbient Existing athenahealth EHR customers. Native, seamless integration with athenaOne EHR. Included at no additional cost for athenahealth subscribers. Strictly limited to athenahealth EHR users; still in rollout phase. Abridge Specialty practices and physician groups. Copy-paste workflow; API integrations are in development. Excellent physician-friendly UI; supports patient-facing recordings. Mid-to-high pricing; limited deep EHR integration at present. Suki Assistant Tech-forward clinics seeking workflow automation. Active partnerships and varying integration levels with major EHRs. Voice commands for EHR navigation; automated coding suggestions. Steeper learning curve; higher pricing for full feature set. DeepScribe Small-to-mid-size practices prioritizing value. Standard copy-paste workflows and common EHR integrations. Simple setup and onboarding; competitive subscription pricing. Fewer advanced features; limited specialty-specific optimization. Conclusions Managing Shadow AI requires healthcare systems to move away from static, reactive blocking strategies toward dynamic, data-centric governance. Unvetted consumer-grade AI tools introduce safety, financial, and regulatory liabilities. However, the systemic administrative burden on clinicians makes the adoption of these efficiency-maximizing technologies inevitable. To protect patient privacy and clinical integrity, healthcare delivery networks must deploy multi-layered detection architectures, integrate Data Security Posture Management with real-time Data Loss Prevention, and proactively transition providers to secure, enterprise-sanctioned clinical AI solutions. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

The Regulatory Realignment of Omnibus VII: Implications for European HealthTech and MedTech The adoption of the "Omnibus VII" legislative package by the Council of the European Union on June 29th, 2026, marks a pivotal juncture in the European Union’s digital governance framework. Driven by strategic evaluations of European competitiveness, principally the landmark reports by former European Central Bank President Mario Draghi and former Italian Prime Minister Enrico Letta, this legislative package executes a targeted simplification agenda designed to reduce administrative duplication. For industries operating at the high stakes intersection of digital technology and healthcare, this package attempts to resolve a core systemic tension: the imperative to foster rapid clinical innovation versus the necessity of maintaining robust fundamental rights and safety safeguards. Prior to the introduction of Omnibus VII, the rapid implementation of the European Union Artificial Intelligence Act (Regulation (EU) 2024/1689) had provoked widespread warning from the life sciences and medical technology sectors. Industry advocates argued that layering horizontal, uncoordinated AI rules on top of the existing, highly demanding Medical Devices Regulation (Regulation (EU) 2017/745, or MDR) and In Vitro Diagnostic Medical Devices Regulation (Regulation (EU) 2017/746, or IVDR) would paralyze digital health innovation. The risk of duplicative audits, separate technical documentation structures, conflicting risk-mitigation standards, and severe shortages in Notified Body capacity threatened to further delay patient access to life-saving technologies. By extending transition timelines, introducing regulatory flexibility, and streamlining institutional pathways under the Cyprus presidency, Omnibus VII represents a pragmatic phase of EU AI governance. However, the refusal of EU co-legislators to grant medical devices a complete sectoral exemption has left the healthtech industry in a complex regulatory position, facing parallel compliance obligations that demand careful, long-term strategic planning. Structural Postponements and Transition Timelines Crucially, Omnibus VII does not rewrite the core principles of the AI Act; rather, it radically recalibrates the enforcement timeline to give developers, national competent authorities, and conformity assessment bodies necessary preparation time. Recognizing that crucial harmonized standards and conformity infrastructures were not yet mature, the co-legislators fast-tracked amendments to postpone key compliance deadlines. The revised timeline distinguishes between stand-alone high-risk AI systems (governed by Annex III of the AI Act) and high-risk AI systems embedded as safety components in products covered by sectoral harmonization legislation (governed by Annex I, which includes medical devices and diagnostics). AI System Category or Legislative Milestone Original Application Date Revised Application Date Transition Window & Strategic Purpose Stand-alone High-Risk AI Systems (Annex III / Article 6(2)) August 2, 2026 December 2, 2027 [cite: 1, 2, 14] 16-Month Extension to finalize national supervisory infrastructures and wait for European harmonized standards. Embedded High-Risk AI Systems (Annex I / Article 6(1), including MDR/IVDR) August 2, 2026 August 2, 2028 [cite: 1, 2, 14] 24-Month Extension to align AI requirements with ongoing targeted revisions of the MDR and IVDR frameworks. AI Regulatory Sandboxes(National Level) August 2, 2026 August 2, 2027 [cite: 2, 3, 6] 12-Month Postponement allowing national competent authorities to build out operational testing environments. Watermarking of Synthetic AI Content (Article 50(2)) August 2, 2026 December 2, 2026 [cite: 13, 14, 15] 4-Month Extension for synthetic content providers, though the post-market implementation grace period was compressed from 6 to 3 months. Prohibitions on Non-Consensual Intimate Content (NCII) / CSAM N/A December 2, 2026 [cite: 13, 14, 15] Immediate Ban targeted at "nudifier" applications and non-consensual deepfakes, taking effect late 2026. This phased implementation provides significant operational relief. However, manufacturers must recognize that some obligations under the AI Act remain active or have already taken effect. For example, the mandatory AI literacy obligation for providers and deployers of AI systems has applied since February 2025, although Omnibus VII softened its terms from an open-ended mandate to a softer requirement to "take measures to support" staff literacy. Furthermore, key transparency obligations under Article 50, including the mandate to disclose when users are interacting with an AI system, such as a patient-facing triage chatbot, remained locked to their original enforcement date of August 2, 2026. The Sectoral Integration Debate: Horizontal Safeguards versus Sector-Specific Pathways The most contentious debate during the negotiation of Omnibus VII centered on how the AI Act should interact with sector-specific product safety regimes. For the life sciences and medical technology sectors, this debate evolved into a legislative clash between two distinct directorates of the European Commission, representing diverging philosophies of regulatory simplification. The first philosophy, championed by the Directorate-General for Communications Networks, Content and Technology (DG CONNECT), was embodied in the "Digital Omnibus" proposal. This approach sought to preserve the horizontal integrity of the AI Act by maintaining systematic AI Act safeguards over medical technologies by default, while introducing administrative coordination mechanisms to streamline compliance. The second, more radical philosophy was proposed in December 2025 by the Directorate-General for Health and Food Safety (DG SANTE) as part of a sweeping initiative to simplify the MDR and IVDR. DG SANTE argued for a complete sectoral carve-out, proposing to move the MDR and IVDR from Section A of Annex I (which triggers the direct application of substantive high-risk AI Act requirements) to Section B. Under the Section B model, the substantive requirements of the AI Act would cease to apply directly to medical devices. Instead, the MDR and IVDR would serve as the sole, primary rulebook for medical AI, with the Commission retaining the power to introduce specific AI requirements later via delegated or implementing acts. During the May 2026 trilogue negotiations, a stark division emerged. Under intense pressure from member states such as Germany, industrial machinery manufacturers successfully secured the Section B shift, moving the Machinery Regulation from Section A to Section B. For a typical manufacturing SME, this carve-out eliminated up to €600,000 in duplicative Year 1 compliance costs. However, the Council of the European Union rejected a similar carve-out for medical technologies. Consequently, medical devices and IVDs were "left behind," remaining under Section A and subject to the full weight of parallel compliance under both the AI Act and the MDR/IVDR. To soften this regulatory burden, negotiators agreed on a compromise mechanism. Rather than a blanket exemption, the European Commission is empowered to adopt implementing acts to limit the application of specific AI Act requirements where the MDR or IVDR is demonstrated to contain equivalent safeguards. While this "equivalence mechanism" provides a legal pathway to reduce duplication, it introduces transitional uncertainty, as the industry must wait for the Commission to draft and adopt these implementing acts. Regulatory Dimension DG CONNECT "Digital Omnibus" (Trilogue Outcome) DG SANTE "MDR/IVDR Simplification" Proposal Annex I Classification Maintained under Section A (direct applicability of horizontal AI Act requirements). Proposed shift to Section B (exempting devices from direct substantive AI Act obligations). Primary Rulebook Parallel, overlapping application of the AI Act and MDR/IVDR. Unified sectoral framework; MDR/IVDR acts as the sole primary rulebook for conformity. Conformity Assessments Streamlined procedures but keeping parallel legislative evaluations. Single conformity assessment pathway strictly embedded in existing MDR/IVDR processes. Legislative Oversight Joint authority cooperation with a compromise "equivalence mechanism". Commission delegated powers to write specific AI rules under the MDR/IVDR framework as needed. Legislative Status Formally adopted as part of the Omnibus VII agreement. Rejected by the Parliament and Council during trilogue negotiations. Operational Reality and Compliance Imperatives for SaMD Developers The failure to achieve a complete sectoral carve-out means that developers of Software as a Medical Device (SaMD) and AI-enabled hardware must navigate two highly demanding regulatory frameworks simultaneously. Because the AI Act classifies any AI-enabled device requiring a third-party conformity assessment as high-risk, this dual compliance regime applies to Class IIa, IIb, and III devices under the MDR, and the vast majority of diagnostics under the IVDR. This dual-framework structure creates substantial friction because of misaligned definitions and regulatory philosophies. For example, the AI Act requires developers to minimize algorithmic and operational risks "as far as technically feasible," whereas the MDR relies on a "benefit-risk balancing" approach. Furthermore, the AI Act introduces the concept of "substantial modification", which can trigger entirely new conformity assessments for self-learning algorithms—while the MDR relies on the distinct concept of "significant change". These misalignments create deep lifecycle management uncertainties. The overall financial impact is substantial. Independent assessments indicate that the AI Act could cost the European economy up to €31 billion over five years, leading to an estimated 20% contraction in AI investment. In April 2026, OpenEvidence, a generative AI clinical decision-support platform utilised by approximately 42% of physicians in the United States, withdrew its services from the European market, citing the impossibility of meeting the AI Act's high-risk compliance hurdles under its current operational model. Similar compliance concerns have slowed the deployment of clinical AI scribes, predictive diagnostic tools, and automated imaging assistants across European clinical networks. To mitigate these bottlenecks, Omnibus VII introduces a unified designation pathway for conformity assessment bodies. Under the new Article 29(4), independent Notified Bodies can submit a single application and undergo a unified assessment procedure to obtain joint designation under both the AI Act and the MDR/IVDR. This administrative streamlining aims to accelerate the availability of AI-competent Notified Bodies, helping to prevent the severe certification bottlenecks that delayed MDR implementation. High-Risk AI Act Expectation Corresponding MDR/IVDR Requirement Integrated Compliance Strategy under Omnibus VII Conformity Assessment Annex IX/X/XI third-party audits by a designated Notified Body. Leverage the unified application pathway to select a Notified Body with joint AI Act and MDR/IVDR credentials. Risk Management System ISO 14971 continuous risk management across the product lifecycle. Embed AI-specific risk profiles (e.g., automation bias, model drift) directly into the existing ISO 14971 file. Technical Documentation Annex II and III comprehensive technical files proving safety and performance. Consolidate documentation into a single technical file under the AI Act Article 11 / Annex II integration allowance. Cybersecurity Controls Annex I General Safety and Performance Requirements (GSPR) on software security. Leverage the Cyber Resilience Act alignment: compliance with CRA Article 12 satisfies AI Act Article 15. Post-Market Surveillance Active Post-Market Clinical Follow-up (PMCF) and periodic safety reporting. Integrate AI drift tracking into PMCF; benefit from the removal of the rigid, standalone AI Act PM plan. The Concurrent Overhaul of the MDR and IVDR As European medical technology companies grapple with the direct application of the AI Act, they must simultaneously navigate a sweeping parallel overhaul of the underlying MDR and IVDR frameworks. The European Commission’s simplification proposals, expected to reach formal adoption between summer 2026 and mid-2027, represent a massive effort to streamline product certification, alleviate chronic device shortages, and lower barriers to entry for smaller developers. This health-sector reform introduces fundamental changes that significantly lighten the administrative load. Key among these is the relaxation of the Person Responsible for Regulatory Compliance (PRRC) requirements. Under the original MDR, small and micro-enterprises were forced to maintain a PRRC "permanently and continuously". The upcoming amendments soften this standard, requiring only that the PRRC be "available". Additionally, health institutions will gain the flexibility to share "home brew" in-house in vitro diagnostics with other legally independent hospitals if it serves public health, eliminating the previous restriction that limited such devices only to cases where no market alternative existed. Crucially, the update also lowers the administrative reporting burden. The frequency for updating Periodic Safety Update Reports (PSUR) has been cut in half: updates are required only every two years—rather than annually—for Class IIb/III medical devices and Class C/D IVDs, and "as necessary" for Class IIa systems. Manufacturers will also benefit from targeted adaptations of classification rules, allowing certain reusable surgical instruments, active implant accessories, and specific clinical software algorithms to be reclassified into lower-risk tiers. Furthermore, to help alleviate the financial strain on early-stage innovators, Notified Bodies will be legally mandated to apply substantial fee discounts: at least a 50% reduction for micro-enterprises, 25% for small enterprises, and 50% for developers of orphan medical devices. To bring commercial predictability to the notoriously slow CE-marking process, the European Commission adopted Implementing Regulation 2026/977 on May 4, 2026. This regulation establishes strict maximum assessment timelines for Notified Bodies, creating a standardised operational rhythm. Conformity Assessment Activity Maximum Permitted Timeline under Regulation 2026/977 Initial Application Review 30 Days Quality Management System (QMS) Audit 120 Days Product Design and Verification Assessment 90 Days Final Certificate Issuance 20 Days While these strict caps do not apply to existing contracts signed before February 25th, 2027, or to recertifications expiring before November 25th, 2027, they provide a long-awaited framework for scheduling and commercial planning. Meanwhile, the global landscape is fragmenting. In the United Kingdom, the government has published its draft Medical Devices (Amendment) Regulations 2026, targeting a December 2026 adoption. This framework proposes a risk-proportionate classification system and introduces an international reliance pathway. Under this reliance model, UK approved bodies can rely on pre-market approvals already granted by comparable regulators in the United States (FDA), Canada (Health Canada), and Australia (TGA) to grant market access. For European healthtech companies, this creates a stark operational contrast: while the EU remains anchored in a parallel-compliance model under Section A of the AI Act, the UK is leveraging unilateral recognition pathways to accelerate market entry and secure its supply chain. The Regulatory Realignment of Omnibus VII: Implications for European HealthTech and MedTech Interoperability with Other Digital Regulations The simplified AI rules under Omnibus VII must also be evaluated within the broader context of the EU’s "Digital Omnibus" regulation, which simultaneously amends a constellation of data protection, cybersecurity, and electronic communications laws. The Digital Omnibus introduces essential alignment between the General Data Protection Regulation (GDPR) and the NIS2 Directive on cybersecurity. Historically, developers faced different reporting timelines and documentation requirements when a personal data breach also qualified as a critical cybersecurity incident. The new rules align these reporting structures, establishing a single, coordinated notification timeline that dramatically reduces administrative duplication. Furthermore, the Digital Omnibus amends the Data Act to provide enhanced protections for proprietary trade secrets, particularly in scenarios where public authorities request access to medical or connected-device data during emergencies. Legacy medical technologies are also explicitly exempted from the Data Act’s strict data-access obligations, protecting historical device portfolios from retroactive and costly engineering overhauls. Additionally, the regulation introduces a revised definition of personal data under the GDPR: pseudonymised data will no longer be classified as personal data if the holder lacks a reasonable, realistic path to re-identify the underlying individuals. This change lowers the regulatory burden for training, validating, and testing healthcare AI models. For developers of connected medical devices, the package also provides essential clarifications regarding the Radio Equipment Directive. Historically, there was significant concern that standard consumer products incorporating basic AI-driven safety elements, such as a smart cooktop using AI to distinguish between a finger and water droplets, would trigger full, high-risk AI Act compliance assessments. By clarifying that AI components that merely assist users or optimize non-safety performance are excluded, the co-legislators have protected standard connected devices from accidental high-risk classifications. The Ethical, Clinical and Civil Backlash While the business community has welcomed the postponement of timelines and the reduction of compliance burdens, the regulatory relaxations under the Omnibus VII package have triggered sharp resistance from civil society and clinical organizations. The decision to delay binding compliance requirements for high-risk systems to late 2027 and mid-2028 has drawn intense criticism. A coalition of more than 127 civil society organizations, including European Digital Rights (EDRi), Amnesty International EU, the European Center for Not-for-Profit Law (ECNL), the European Disability Forum (EDF), and the European Network Against Racism (ENAR), has condemned the proposal as the largest rollback of digital fundamental rights safeguards in the Union's history. They argue that delaying these strict rules leaves vulnerable populations exposed to unvetted, potentially biased algorithmic decision-making in critical areas like emergency triage, public health resource allocation, and clinical diagnostics. These concerns are echoed by the Standing Committee of European Doctors (CPME), which has raised alarms over the medical privacy implications of the package. CPME’s opposition focuses on the new Article 4a, which allows developers to bypass the standard GDPR Article 9 prohibition and process sensitive medical histories for bias detection and correction. CPME argues that this "deregulatory" shift risks eroding patient trust and violating medical confidentiality. They caution that the expanded use of pseudonymized or anonymised health data under the revised definition of personal data could allow commercial developers to train clinical algorithms without explicit patient consent, potentially leading to unauthorized data reuse and undermining the doctor-patient relationship. Clinicians also emphasize that clinical AI solutions with valid CE markings can yield widely varying results in real-world environments. For example, minor software updates in AI diagnostics can introduce performance drift, potentially leading to false positives or missed diagnoses. Consequently, clinical groups argue that weakening horizontal oversight in favour of slower, sector-specific MDR/IVDR processes could expose patients to algorithmic errors, emphasising that human clinical oversight must remain the final safeguard in patient care. Strategic Playbook for Healthcare AI and MedTech Executives Despite the extended timelines, the core requirement of parallel compliance under both the AI Act and the MDR/IVDR remains a reality. Healthtech executives and regulatory compliance officers must utilize this newly granted transition window to build unified, highly efficient compliance frameworks. Step 1: Execute a Dual-Classification Audit Manufacturers must immediately classify all software and hardware models under both regimes. The key is evaluating clinical algorithms against the newly clarified Article 3(14) "safety function" exemptions. If a software module merely assists a clinician or optimises administrative workflow without autonomously making diagnostic or treatment decisions, it should be documented as falling outside the high-risk classification, avoiding duplicative conformity assessments. Step 2: Navigate High-Risk Exemptions and Register Under the provisional agreement, even if a provider believes their AI system is exempted from high-risk requirements under Article 6(3) of the AI Act, they are still subject to a mandatory registration requirement. This means companies must register these systems in the central EU database for high-risk AI, ensuring full regulatory transparency even when claiming an exemption. Step 3: Establish a Integrated Quality Management System (QMS) Rather than maintaining separate, parallel compliance tracks, developers should integrate AI Act quality management rules directly into their existing ISO 13485 structures. Risk management protocols under the AI Act can be merged directly into the ISO 14971 risk files. Crucially, cyber compliance can be streamlined by leveraging the Cyber Resilience Act alignment: proving compliance with CRA Article 12 automatically satisfies the cybersecurity requirements of AI Act Article 15. Step 4: Leverage Sandbox and Testing Opportunities SMCs and early-stage startups should actively seek access to the newly established Union-level AI sandboxes, managed directly by the European AI Office. Furthermore, developers should exploit the expanded allowance for "real-world testing" under Article 60. This allows high-risk medical AI systems to be tested under real-world clinical conditions before formal market placement, providing valuable clinical performance data while accelerating the path to certification. Step 5: Engage Dual-Designated Notified Bodies When scheduling conformity assessments, manufacturers should specifically target Notified Bodies that have utilised the unified application pathway under Article 29(4). Engaging an auditor designated under both the AI Act and the MDR/IVDR allows for a single, integrated audit process. This eliminates duplicative testing and significantly reduces overall certification costs. Conclusion and Future Outlook The simplified AI rules under Omnibus VII provide a critical transition window for the European healthtech and medtech sectors, but they do not resolve the structural challenges of dual regulation. While the extension of compliance deadlines to August 2028 provides immediate breathing room, companies must recognise that this delay is a preparation window, not a deregulation of medical AI. The long-term regulatory environment will be heavily shaped by two key factors. First is the European Commission’s use of its newly granted powers to adopt implementing acts to limit AI Act requirements where equivalent MDR/IVDR safeguards exist. Second, the ongoing targeted revisions to the MDR and IVDR represent the final, critical opportunity for the industry to lobby for complete sectoral integration. If these revisions successfully embed AI-specific safeguards directly into the medical device regulations, Europe can establish a unified, predictable, and competitive framework for medical AI. Until then, healthtech executives must proactively integrate their compliance systems, leveraging the transition period to build robust, scalable, and audit-ready products. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

The Nelson Advisors Guide for Founders to HealthTech and MedTech Success in 2026: 10 European Case Studies Executive summary Europe has quietly become one of the most productive regions on earth for building health and medical technology companies of genuine scale. Over the past decade a cohort of founders has turned the continent's structural disadvantages, fragmented markets, multiple regulators, conservative public payers, slow procurement, into the very foundations of durable, defendable businesses. This guide examines ten of them in depth. The companies profiled here are deliberately varied. They span pure software platforms (Doctolib, Ada Health, Kry/Livi), deep-science instrumentation (Oxford Nanopore), surgical robotics (CMR Surgical), applied genomics and AI biology (SOPHiA GENETICS, Owkin), clinical-AI infrastructure (Corti), tech-enabled care delivery (Cera) and preventive consumer medicine (Neko Health). Between them they have raised well over five billion euros from grants, venture capital, private equity, strategic pharmaceutical partners and the public markets. Some are now profitable; several are not; one cautionary comparator in this report went bankrupt after a multi-billion-dollar valuation. Three themes run through every case. First, the most defendable moats in European health technology are built not from clever algorithms but from proprietary data, regulatory clearance and entrenchment in the workflows of clinicians and payers, assets that take years to accumulate and are extremely hard to copy. Second, the distinction between being “AI-native” and an “AI enabler” matters far less than founders assume: what matters is whether AI is layered on top of a genuine distribution and data advantage. Third, the 2022–24 funding winter separated the survivors from the casualties almost entirely on the basis of capital discipline and proximity to whoever actually pays for healthcare. This is not a ranking. It is a working reference for founders and boards: a set of strategies, tactics, funding journeys, critical decisions and hard-won lessons drawn from companies that have already walked the path you are on. Read it as a map of what has worked, what has nearly killed otherwise excellent businesses, and where the genuinely defensible value in this market sits in 2026. How to read this guide ▸ Each case study opens with a fact box (HQ, founders, funding, valuation, AI posture) followed by analysis of strategy, scalability, moat, risks, critical decisions and lessons. ▸ Funding figures are drawn from company announcements and reputable reporting to mid-2026; private valuations are last-round marks and may be stale. Treat them as directional, not audited. ▸ Cross-cutting chapters after the case studies synthesise the patterns: AI-native versus AI-enabler, moat construction, the funding map, and the risks that have sunk well-capitalised peers. The 2026 landscape: why these lessons matter now Founders entering HealthTech and MedTech in 2026 face a market that looks superficially similar to 2021 but behaves completely differently. The exuberant capital of the pandemic era has gone; in its place is a more demanding investor base that wants evidence of clinical impact, a credible path to profitability and a defensible position before it writes large cheques. At the same time, the underlying demand has never been stronger. Ageing populations, chronic disease burden, workforce shortages across European health systems and the maturation of artificial intelligence have combined to make health one of the few categories where technology can simultaneously cut cost and improve outcomes. The opportunity is therefore real, but the bar is higher. The companies in this report succeeded not because they rode a hype cycle but because they solved a structural problem for a payer, a clinician or a patient and then made that solution progressively harder to displace. Understanding how they did it and where comparable companies failed, is the single most useful preparation a founder can do. A simple framework: scalable, sustainable, defendable Throughout this guide we assess each company against three properties that, in our advisory work, separate enduring health-technology businesses from those that raise large rounds and then stall. • Scalable — can the business add revenue faster than it adds cost? Software platforms scale through near-zero marginal cost; hardware and care businesses scale through repeatable unit models, manufacturing capacity or acquisition roll-ups. The question is always: what is the unit, and does the next unit get cheaper? • Sustainable — does the business have a route to funding itself? In a market where reimbursement is slow and capital is expensive, the survivors are those with recurring revenue, demonstrated unit economics and the discipline to reach profitability before the money runs out. • Defendable — once the business works, how hard is it to copy? In health the durable moats are proprietary longitudinal data, regulatory clearance, entrenchment in clinical workflows, patent estates and trusted brand. Features are not moats; the assets that take years to build are. With that framework in mind, we turn to the ten companies. Why these ten companies These ten were not chosen as a definitive ranking of the best European health-technology companies, nor as an investment recommendation. They were selected because, between them, they illustrate the full range of strategies, business models, funding paths and moats a founder is likely to consider, also because each has progressed far enough to offer real, tested lessons rather than early promise. The selection deliberately spans the spectrum from pure software to deep science. Doctolib, Ada Health and Kry/Livi represent software-led platforms and digital care. Oxford Nanopore and CMR Surgical represent capital-intensive deep-science hardware. SOPHiA GENETICS and Owkin sit at the data-and-AI layer of genomics and drug discovery. Corti represents pure clinical-AI infrastructure. Cera represents technology-enabled care delivery, and Neko Health represents consumer-facing preventive medicine. The set also spans the funding journey, from companies still private and venture-backed, through those that took strategic capital from industry partners, to two that have weathered the public markets — and several European geographies, with companies anchored in France, the United Kingdom, Switzerland, Germany, Sweden and Denmark. Just as importantly, the cohort spans outcomes. Some are profitable; several are not; valuations have risen for some and fallen sharply for others. We have resisted the temptation to present only unambiguous successes, because the most useful lessons frequently lie in the tensions, a brilliant business with a punished share price, a category creator still searching for profitability, an AI pioneer confronting a new paradigm. Read together, they form a more honest map of the territory than a list of winners would. Part One: Ten European case studies The ten companies are grouped loosely from broad software platforms through to deep science and consumer medicine. Each can be read independently. 1. Doctolib — the operating system for European healthcare HQ / country Paris, France (major hub in Berlin) Founded 2013 Founders Stanislas Niox-Château (CEO) and co-founders Sub-sector Practice-management SaaS, e-booking and teleconsultation Funding to date ~€790m+ private; never IPO’d (Series F €500m, 2022) Key backers Accel, General Atlantic, Eurazeo, Bpifrance Valuation ~€5.8bn / $6.4bn (2022, reportedly held flat in 2025) AI posture AI enabler — marketplace first, AI layered on from 2024 What it does and why it matters Doctolib is the closest thing Europe has to an operating system for healthcare. Patients use it to find a practitioner, book an appointment and run a video consultation; the much more valuable side of the business sells practice-management software, scheduling, records, billing and, increasingly, AI tools, to doctors, clinics and hospitals. Founded in Paris in 2013 by Stanislas Niox-Château, it now serves more than 135,000 healthcare professionals and is the most valuable startup France has produced. Its significance for founders is that it demonstrates how to build a category-defining platform in a regulated, fragmented, trust-sensitive market and how distribution, not technology, became the moat. Strategy and tactics Doctolib executed a textbook land-and-expand vertical-SaaS strategy. It began with the single sharp problem of appointment booking, then layered teleconsultation, full practice management, payments and AI on top, increasing revenue per practitioner over time. Crucially, growth was driven by a large human field-sales force that built density city by city, signing up doctors in person to overcome the trust barrier that gates technology adoption in medicine. This created a two-sided flywheel: more doctors made the platform more useful to patients, and more patient demand pulled in more doctors. The company’s decision to run COVID-19 vaccination booking for the French state in 2020–21 cemented both its brand and its status as quasi-national infrastructure. Scalable, sustainable, defendable The model is highly scalable: subscriptions of roughly €109–€229 per practitioner per month carry high gross margins and low marginal cost once local density is achieved. On sustainability, 2024 annual recurring revenue reached €348m, up 22.5%, while losses were cut 38% to €53.8m, putting breakeven within reach. Defensibility comes from per-geography network effects, the high switching costs of being a practice’s system of record, GDPR-compliant health-data hosting, and a deep brand of trust. AI-native versus AI-enabler Doctolib is an AI enabler, not an AI-native company. It was built as a marketplace and bolted AI on from 2024 — an ambient-transcription Consultation Assistant (built on Azure OpenAI and the acquisition of Typeless) and an AI phone assistant in 2025. The instructive point is that its AI is valuable precisely because it sits on top of an existing network, distribution and a decade of appointment and clinical data. AI is the layer; the moat is underneath it. Risks and critical decisions The principal risks are recurring data-privacy and regulatory scrutiny, including controversy over using anonymised patient data to train AI, a continued path to profitability and roughly 80% revenue concentration in France. Three critical decisions defined the company: choosing a heavy human field-sales model over pure product-led growth; stepping up to run national vaccination booking; and the decisive 2024 pivot to position itself as an AI clinical platform ahead of a possible 2026–27 IPO. Founder lessons from Doctolib ▸ In regulated verticals, owning the system of record and the local network beats winning on features. ▸ Boots-on-the-ground sales can itself be a moat when trust gates adoption. ▸ A crisis (COVID) can be a catalyst that converts a product into infrastructure. ▸ AI is a layer on top of distribution and data — not a substitute for them. 2. Oxford Nanopore Technologies — owning a new sequencing category HQ / country Oxford, United Kingdom (LSE: ONT) Founded 2005 (University of Oxford spin-out) Founders Hagan Bayley, Gordon Sanghera, Spike Willcocks Sub-sector Genomics instrumentation and consumables (sequencing) Funding to date >$1.1bn pre-IPO; ~£350m raised at 2021 IPO Key backers IP Group, Invesco, Temasek, Nikon, Wellington Valuation ~£1.2bn market cap (2026), down ~75% from IPO peak AI posture AI-enabled deep tech — ML basecalling on a physics moat What it does and why it matters Oxford Nanopore sells a fundamentally different way to read genetic material. Its sequencers pass single DNA or RNA molecules through protein nanopores and read the resulting electrical signal in real time, enabling portable, long-read sequencing, from the USB-stick-sized MinION to the high-throughput PromethION. Spun out of the University of Oxford in 2005 around Professor Hagan Bayley’s chemistry, it created and still leads a sequencing category distinct from the dominant short-read, sequencing-by-synthesis approach. It matters to founders as the archetype of patient, deep-science company building: more than $1.1Bn raised over roughly sixteen years before a landmark IPO, and a moat rooted in physics, biochemistry and intellectual property rather than software alone. Strategy and tactics The strategy was to build one scalable sensing platform, “analysis of anything, by anyone, anywhere” and ladder products from the entry-level Flongle up to PromethION. Tactics included democratising access and seeding academic users with grant programmes (the platform now underpins more than 14,000 publications), continuous accuracy improvements, and a deliberate move up into higher-margin Clinical and BioPharma segments. From 2025 the company showed new discipline, sunsetting weaker products to focus on its profitability path. Scalable, sustainable, defendable Scalability comes from a common sensor across the entire product line and a razor-and-razor-blade model: consumable flow cells and reagents generate recurring revenue tied to the installed base. FY2025 revenue reached £223.9m, up around 24% at constant currency, with Clinical growing roughly 60%. With about £302m of cash at year-end 2025 and guidance for adjusted EBITDA breakeven in 2027, the business is funding its own path to sustainability. Defensibility rests on a deep, litigated patent estate covering pores, enzymes and electronics. AI-native versus AI-enabler Oxford Nanopore is best described as AI-enabled deep tech. The core invention is physical and biochemical, but its Dorado basecaller uses neural networks to convert raw electrical signal into sequence in real time, and machine learning is now used to design novel pore proteins. AI is mission-critical software, but it sits on a moat made of physics and biology, the inverse of a pure-software AI company. Risks and critical decisions The clearest risk is illustrated by the share price: down roughly 75% from its IPO peak, a reminder that a blockbuster listing is a milestone, not a finish line. The company also has meaningful exposure to US federal research funding and is in litigation with BGI/MGI. Three critical decisions shaped it: spinning out around nanopore science in 2005 as a contrarian long-horizon bet; re-engineering its pore chemistry after Illumina’s 2016 patent suit (switching to a CsgG pore) to defuse the litigation; and choosing a London listing in 2021. Founder lessons from Oxford Nanopore ▸ Own a category rather than fight an incumbent head-on. ▸ Razor-and-blade economics turn hardware sales into recurring revenue. ▸ Patents are a living asset — be prepared to re-engineer around your own IP under litigation. ▸ Deep science needs deep, patient capital; a spectacular IPO is the start of public-market discipline, not the end of the journey. 3. CMR Surgical — challenging a monopoly in surgical robotics HQ / country Cambridge, United Kingdom Founded 2014 (as Cambridge Medical Robotics) Founders Mark Slack, Luke Hares, Martin Frost, Paul Roberts, Keith Marshall Sub-sector Surgical robotics (minimally invasive soft-tissue surgery) Funding to date ~$1.4–$1.5bn private (Series D $600m, 2021) Key backers SoftBank Vision Fund 2, Ally Bridge, Tencent, GE Healthcare Valuation $3bn (2021); explored a sale up to ~$4bn in 2025 AI posture AI enabler — mechatronics core, AI/data features added What it does and why it matters CMR Surgical makes Versius, a modular, portable robotic-assisted surgery system that competes directly with Intuitive Surgical’s dominant da Vinci. Where da Vinci is a single large console-and-cart, Versius uses small, independent bedside arm carts that fit existing operating theatres and can be moved between them. Founded in Cambridge in 2014, Versius has now been used in more than 30,000 procedures across 30-plus countries. It matters because it is a credible European challenger to one of medtech’s most entrenched monopolies — and a live lesson in the capital intensity and patience that hardware in medicine demands. Strategy and tactics The strategy was to win as the challenger by competing on cost, footprint and accessibility rather than trying to out-feature the incumbent. Tactically, CMR commercialised internationally first, across Europe, Latin America, Asia and Australia, becoming the second most-adopted soft-tissue robot outside the US before tackling the American market. It built recurring revenue through single-use instruments, service contracts and its Versius Connect data ecosystem, and invested in its own manufacturing capacity in Ely. Scalable, sustainable, defendable Scalability rests on a razor-and-blade model (system plus single-use instruments plus service) and purpose-built manufacturing capacity of around 500 systems a year. The defensibility is real but shallower than the incumbent’s: surgeon and theatre switching costs, multi-jurisdiction regulatory clearances and a growing procedure dataset, set against Intuitive’s roughly 60% global share, 9,000-plus installed systems and around 85% recurring revenue. CMR’s edge is economics and form factor, not scale. AI-native versus AI-enabler CMR is firmly an AI enabler, not AI-native. The core value is mechatronics, robotics and surgeon ergonomics; the robot is teleoperated, not autonomous. AI and data features, procedure analytics, the digital ecosystem, near-infrared imaging, are add-ons layered onto excellent hardware. Risks and critical decisions The risks are formidable: a dominant incumbent, deep-pocketed new US entrants (Medtronic’s Hugo, J&J’s Ottava), high cash intensity, slow surgical sales cycles, a 2025 reliance on debt and a valuation that has been flat since 2021. Three critical decisions stand out: the founding bet on a modular, portable architecture; the decision to internationalise before entering the US; and new leadership’s high-conviction call to scrap the planned first-generation US launch and wait for the cleared Versius Plus, vindicated by FDA clearance in December 2025. Founder lessons from CMR Surgical ▸ Against an entrenched incumbent, differentiate on architecture and economics — don’t try to out-feature them. ▸ Sequence your markets: prove the product in less-defended geographies before the hardest one. ▸ Build recurring revenue into hardware from day one. ▸ Be willing to delay a flagship launch for the right product — and respect that hardware in medicine is brutally capital-intensive. 4. SOPHiA GENETICS — owning the genomic data layer HQ / country Lausanne, Switzerland / Boston, USA (NASDAQ: SOPH) Founded 2011 (EPFL spin-out) Founders Jurgi Camblong, Pierre Hutter, Lars Steinmetz Sub-sector Genomic data analytics / clinical bioinformatics SaaS Funding to date ~$250m pre-IPO; $234M gross at 2021 NASDAQ IPO Key backers Balderton, Generation IM, aMoon, Hitachi Ventures Valuation ~$330–360M market cap (2026), down ~70% from IPO AI posture AI-native within its niche (proprietary ML is the product) What it does and why it matters SOPHiA GENETICS sells SOPHiA DDM, a cloud-based, AI-powered platform that helps hospitals and laboratories analyse and interpret complex genomic and multimodal data for precision medicine, principally in oncology. It sits one layer above the sequencing hardware companies, turning raw sequencer output into clinically actionable insight. Founded in 2011 as an EPFL spin-out, its platform has analysed more than 2.5 million cases across some 70 countries. It matters as a rare European deep-tech company that listed on NASDAQ, and as a case study in owning the data and analytics layer of a value chain rather than the hardware. Strategy and tactics The strategy is a decentralised “data-as-a-network” model: hospitals sequence locally and analyse in SOPHiA’s cloud, so each analysis enriches a shared dataset that improves the algorithms for everyone. Tactics include classic land-and-expand SaaS economics, licensing marquee assays (notably Memorial Sloan Kettering’s MSK-ACCESS and MSK-IMPACT), a 2026 joint venture with MSK in multimodal precision oncology, and monetising the platform with biopharma partners. Scalable, sustainable, defendable The cloud SaaS model is scalable with improving margins (adjusted gross margin around 74%), and revenue reaccelerated to roughly $77m in FY2025, up about 19%. It is not yet self-sustaining, net losses remain large, with adjusted EBITDA breakeven targeted by end-2026. Defensibility comes from a genuine data network effect, a dataset of more than two million genomic profiles, around 31 patents and deep clinical integration that creates high switching costs. AI-native versus AI-enabler SOPHiA is AI-native within its niche: proprietary machine learning is the product, not an add-on. It is, however, an applied AI company for genomics rather than a frontier-model builder, a useful distinction for founders, because applied, domain-specific AI on proprietary data is often more defensible than generic model capability. Risks and critical decisions Risks include persistent losses, a stock down around 70% from IPO, intellectual-property litigation (a European dispute with Guardant Health) and a leadership transition. Three critical decisions defined it: building a decentralised analytics layer rather than a centralised lab; listing on NASDAQ with a dual Swiss-Boston identity; and going all-in on multimodal oncology through the MSK alliance, alongside a deliberate founder-to-CEO succession in 2026. Founder lessons from SOPHiA GENETICS ▸ Own the data layer, not just the workflow — a data network effect compounds with every customer. ▸ Partner with the prestige incumbent (MSK) for instant clinical credibility. ▸ A great moat does not guarantee a great stock; unit economics must eventually deliver. ▸ Plan IPO timing, venue and founder succession deliberately, not reactively. 5. Owkin — strategic capital and an AI-biology moat HQ / country Paris, France / New York, USA Founded 2016 Founders Thomas Clozel (CEO), Gilles Wainrib (President/CSO) Sub-sector AI for drug discovery, biomarkers and trial optimisation Funding to date ~$300m+ equity (Sanofi $180m equity + $90m alliance, 2021) Key backers Sanofi, Bristol Myers Squibb, GV (Google Ventures), Bpifrance Valuation ~$1bn unicorn (last mark; likely stale) AI posture Clearly AI-native — founded around ML on patient data What it does and why it matters Owkin is a French-American “TechBio” company that applies AI and federated learning to hospital data to discover drug targets and biomarkers, optimise clinical trials and build diagnostics. Federated learning lets it train models across hospital datasets without the data ever leaving the institution, a privacy-preserving approach that unlocked access to data others could not touch. It became a unicorn in 2021. It matters as the clearest example in this report of strategic capital: rather than raising plain venture money, Owkin sold equity to the pharmaceutical giants whose problems it solves. Strategy and tactics The strategy is to secure privacy-preserving access to multimodal hospital data, train biology-specialised AI on it, and monetise through pharma partnerships and its own diagnostics and pipeline. Tactically, the defining move was Sanofi’s November 2021 investment of $180m in equity plus $90m for a three-year R&D alliance, the round that made Owkin a unicorn, followed by a deal with Bristol Myers Squibb. From 2025 the company pivoted toward a more scalable platform and licensing model built around an agentic AI co-pilot and a foundational biology model, and spun out its diagnostics business. Scalable, sustainable, defendable Scalability improves markedly with the shift from bespoke services to a licensable software platform. Sustainability is trending the right way, revenue around $86m and roughly doubling, with about $200m of cash and shrinking losses, though the business remains pre-profit and capital-intensive. Defensibility rests on a decade of hospital trust and data access (including a proprietary multi-omics cancer atlas), federated-learning IP and credibility, and embedded relationships with pharma partners who are also shareholders. AI-native versus AI-enabler Owkin is unambiguously AI-native, it was founded around machine learning on patient data and is now building foundational biological models. For founders this case shows both the upside and the dependency of being AI-native: the technology is the company, which makes the data and partnerships that feed it existential. Risks and critical decisions Risks include admitted over-hiring after its mega-rounds, concentration in a few large partners, a potentially stale billion-dollar valuation (down-round risk), an unproven drug pipeline and a frontier-model arms race against far larger players. Three critical decisions defined it: taking Sanofi’s strategic equity and alliance over pure VC; pivoting from services toward a licensable platform; and spinning out diagnostics to refocus the core. Founder lessons from Owkin ▸ Strategic capital — equity plus data plus credibility from an industry partner — can beat plain VC, but it adds dependency. ▸ A regulated-industry data moat is built slowly and is correspondingly hard to copy. ▸ Be honest about over-hiring after mega-rounds; capital can mask a lack of focus. ▸ Evolve from bespoke services toward scalable software, and divest non-core assets to stay focused. The Nelson Advisors Guide for Founders to HealthTech and MedTech Success in 2026: 10 European Case Studies 6. Ada Health — turning regulation into a moat HQ / country Berlin, Germany Founded 2011 (consumer app launched 2016) Founders Claire Novorol, Martin Hirsch, Daniel Nathrath (CEO) Sub-sector AI symptom assessment, triage and care navigation Funding to date ~$190m+ (Series B $90m, 2021; €30m venture debt 2023) Key backers Access Industries, Leaps by Bayer, Samsung Catalyst Fund Valuation ~$600m last reported (2022); likely repriced lower AI posture AI-native, but pre-LLM expert-system AI What it does and why it matters Ada Health builds an AI-powered symptom assessment, triage and care-navigation platform. A consumer app launched in 2016 lets users describe symptoms and receive a probabilistic assessment and guidance on what to do next; by late 2024 it had handled more than 35 Million assessments. Founded in Berlin (and not to be confused with the unrelated Canadian customer-service company Ada Support), it is one of Europe’s most-used digital triage tools. It matters as a case study in converting regulatory clearance into a durable moat, and in the difficult art of monetising a free consumer health product. Strategy and tactics The central strategic move was a pivot from a free consumer app, which built brand, reach and a vast dataset, to an enterprise and B2B2C model, selling “Ada Assess” to health systems, payers, pharma and governments including Bayer, Novartis, Pfizer and US health systems. After the 2022 downturn the company repositioned hard around profitability and clinical credibility, and used venture debt to extend runway with less dilution. Scalable, sustainable, defendable The product is highly scalable: software-only AI triage with near-zero marginal cost, available in multiple languages. On sustainability, the company stated it reached profitability at the end of 2023 with 260% year-on-year revenue growth, driven by enterprise contracts. Defensibility comes from a physician-curated medical knowledge base, EU-MDR Class IIa certification and ISO 13485, a trusted consumer brand and 35 million-plus structured real-world assessments. AI-native versus AI-enabler Ada is AI-native but in a specific, instructive sense: it was built around a proprietary probabilistic medical-reasoning engine, not a bolt-on, yet that engine predates the generative-LLM wave. Large language models are now both a competitive threat and a strategic question for the company, illustrating that being AI-native at one moment does not guarantee defensibility against the next paradigm shift. Risks and critical decisions Risks include LLM-driven disruption (ChatGPT, Google and others), medical-advice liability, thin consumer monetisation and enterprise concentration. Three critical decisions defined it: the early pivot from a doctor-facing network to a consumer symptom checker; the later pivot to a monetised enterprise model prioritising profitability after 2022; and the early investment in EU-MDR certification, which turned a compliance cost into a competitive moat. Founder lessons from Ada Health ▸ In regulated markets, certification and clinical validation are a durable moat, not just a cost. ▸ A free consumer product can be a data-and-brand engine — but it rarely funds deep tech. Find the enterprise payer. ▸ In a downturn, choose profitability over vanity growth. ▸ Being AI-native at one moment is not permanent defensibility — you must actively renew it against paradigm shifts such as LLMs. 7. Kry / Livi — discipline and the hybrid care model HQ / country Stockholm, Sweden (Livi in UK and France) Founded 2014 Founders Johannes Schildt (CEO), Josefin Landgård and co-founders Sub-sector Digital-first primary care / telehealth (hybrid) Funding to date ~$700m total (Series D ~$300m, 2021); no IPO Key backers Accel, Index Ventures, Ontario Teachers’, CPP Investments, Fidelity Valuation ~$2bn last mark (2021); likely stale post-repricing AI posture AI adopter — clinical-ops company, AI as a margin lever What it does and why it matters Kry, branded Livi in the UK and France, is a digital-first primary care company offering app-based video consultations alongside owned physical clinics, a hybrid “phygital” model across Northern Europe. Founded in Stockholm in 2014, it now serves more than seven million patients with over 1,300 clinicians. It matters above all as a survival story: Kry came through the telehealth crash that bankrupted Babylon Health, and did so through capital discipline rather than momentum. Strategy and tactics The strategy combined digital and physical care, the app as a funnel, owned clinics for higher-value and chronic care, with multi-country expansion built on genuine localisation rather than copy-paste (per-visit payment in Sweden versus free-at-point-of-use in the NHS). Tactics included expanding wallet share into mental health, weight management and occupational health, acquiring clinics, and anchoring revenue to public payers with a cost-saving argument of roughly 40% versus in-person care. Scalable, sustainable, defendable The model proved scalable, the patient base grew roughly a hundredfold in five years and AI now cuts clinician administrative time by around 30%. On sustainability, all of its markets were individually profitable by end-2024, with group profit deliberately delayed to fund generative-AI investment; FY2024 revenue was around €220m, up 13%. Defensibility comes from an employed clinician network (quality control versus US-style marketplaces), public-payer and government contracts, multi-country regulatory know-how and owned clinic assets. AI-native versus AI-enabler Kry is an AI adopter, not AI-native. It was built as a telehealth and clinical-operations company, with AI layered in for efficiency. Tellingly, its CEO has publicly framed generative AI as an overhead that is delaying group profitability, a candid view of AI as a margin lever rather than the core product. Risks and critical decisions Risks include the commoditisation of video consultation (now a feature inside larger platforms such as Doctolib), roughly 78% revenue concentration in Sweden, and dependence on reimbursement policy. Three critical decisions defined it: the 2022 pivot from hyper-growth to profitability, cutting around 400 jobs and exiting the German consumer market — which arguably saved the company; the hybrid owned-clinic strategy; and the deliberate decision to delay group profitability to fund AI. Founder lessons from Kry / Livi ▸ Multi-country health expansion needs real localisation, not copy-paste. ▸ Discipline beats momentum in a downturn — Kry survived where Babylon failed. ▸ Own the hard, defensible parts (employed clinicians, clinics), and anchor to whoever actually pays. ▸ Treat AI as a margin lever and be transparent with investors about the trade-off. 8. Cera — a data moat from owning care delivery HQ / country London, United Kingdom Founded 2015–16 Founders Dr Ben Maruthappu (CEO), Marek Sacha Sub-sector Digital-first home healthcare / care delivery Funding to date ~$571m (£450m) across ~13 rounds; latest >$150m, 2025 Key backers Kairos, BDT & MSD Partners, Schroders Capital, Vanderbilt Endowment Valuation Unicorn (>$1bn) as of 2025; exact figure undisclosed AI posture AI enabler evolving to AI-led — care operator first What it does and why it matters Cera is Europe’s largest digital-first home healthcare provider, delivering AI-powered care, nursing and telehealth in patients’ homes as an alternative to hospital and residential care. Founded in 2015–16 by NHS doctor Ben Maruthappu after his mother received fragmented post-fracture care, it employs more than 10,000 carers and nurses, works with over 150 local authorities and around two-thirds of NHS Integrated Care Systems. It matters because, unusually for the sector, Cera is profitable, and because its decision to own care delivery created a proprietary data moat that pure-software rivals cannot replicate. Strategy and tactics The strategy is to replace expensive hospital and residential beds with technology-enabled care at home. Cera built an actual care-delivery business, it employs carers and layered a proprietary digital platform on top, using AI to predict deterioration and falls from daily carer-collected data and intervene early (it claims to predict up to around 80% of deterioration and cut hospitalisations by roughly a third). It grows by acquiring and rolling up fragmented home-care agencies and anchors revenue on recurring public-sector contracts. Scalable, sustainable, defendable Scalability comes from combining software with an acquisition roll-up across a highly fragmented market. On sustainability, Cera reported being EBITDA-positive in 2023 and free-cash-flow positive in 2024, exceptionally rare for a healthtech scale-up, which de-risked its 2025 unicorn round of more than $150M. Defensibility rests on the proprietary longitudinal dataset generated by millions of daily home visits, deep entrenchment with NHS and local-authority commissioners, and scale as the UK’s largest operator. AI-native versus AI-enabler Cera is an AI enabler evolving toward AI-led. Its core business is physical care delivery; AI is layered on to optimise it through predictive risk, scheduling and documentation. It markets itself as “AI-led home healthcare,” but the underlying model is operational first and algorithmic second and that operational layer is precisely what generates the data the AI needs. Risks and critical decisions Risks include heavy dependence on public-sector budgets and contract renewals, margin pressure in low-margin care, labour intensity and wage inflation, scrutiny of its AI claims, and integration risk from acquisitions. Three critical decisions defined it: building a care operator rather than a pure software platform; anchoring on public-sector contracts for defensive recurring revenue; and prioritising profitability over pure growth, which gave it negotiating leverage and resilience. Founder lessons from Cera ▸ Owning the full delivery stack creates a proprietary data moat that software-only rivals lack. ▸ In healthcare, demonstrable cost savings to the payer is the strongest sales argument. ▸ Reaching profitability buys leverage and resilience in a sector littered with cash-burning failures. ▸ Roll-up of a fragmented market can be a legitimate scaling engine — if integration is disciplined. 9. Corti — AI infrastructure for healthcare HQ / country Copenhagen, Denmark Founded 2014–16 Founders Andreas Cleve (CEO), Lars Maaløe (CTO) and co-founders Sub-sector Clinical-AI infrastructure / ambient clinical intelligence Funding to date ~$103m (Series B $60m, 2023) Key backers Prosus Ventures, Atomico, Eurazeo, EIFO Valuation Undisclosed AI posture Genuinely AI-native — healthcare-specific foundation models What it does and why it matters Corti builds healthcare-specialised AI infrastructure, foundation models and APIs for speech-to-text, text generation, agentic workflows and clinical documentation, that act as a real-time co-pilot during clinical conversations. It first became known for AI that detects cardiac arrest during emergency calls. Founded in Copenhagen, its technology now powers applications serving more than 100 Million patients a year and over a million interactions a week, including in the NHS. It matters as one of the clearest AI-native companies in this report, and as a model for scaling through infrastructure rather than selling a single product. Strategy and tactics Corti’s defining strategic move was to pivot from selling a finished co-pilot to becoming AI infrastructure for healthcare developers, opening its API to the world in 2025 and launching healthcare-specific foundation models trained exclusively on healthcare data, explicitly positioned against general-purpose LLMs as safer and more suitable for clinical use. Tactics include an API and SDK platform play, partnerships with EHR vendors and virtual-care platforms, and aggressive US expansion. Scalable, sustainable, defendable The platform model is highly scalable: others build on Corti, multiplying its reach (it reported a roughly fortyfold increase in projects built on its platform over two quarters). Sustainability is less proven, reported revenue of around $13M in 2025 is modest against the capital raised, so the infrastructure pivot must monetise. Defensibility comes from nearly a decade of peer-reviewed research and proprietary clinical training data drawn from millions of real patient interactions, a safety and trust advantage in a domain where general models are risky. AI-native versus AI-enabler Corti is genuinely AI-native: the entire company is built around proprietary healthcare AI models, and AI is the product rather than an add-on. It is the strongest example in this report of a company whose moat is domain-specific AI, although, importantly, the moat is the proprietary clinical data and published evidence, not the model architecture itself. Risks and critical decisions Risks include intense competition from well-funded rivals (Microsoft/Nuance DAX, Abridge, Suki, Nabla) and from general-purpose LLM providers moving into healthcare, clinical-safety and liability burdens, long enterprise sales cycles and modest revenue relative to capital. Three critical decisions defined it: the life-or-death origin in cardiac-arrest detection that built clinical credibility and data; the 2024–25 pivot to an open API and foundation-model business; and the deliberate choice to build healthcare-specific models rather than fine-tune general LLMs. Founder lessons from Corti ▸ In clinical AI, proprietary domain data and published evidence are the moat — not the model architecture. ▸ Specialisation can beat general-purpose incumbents on safety and trust. ▸ Becoming infrastructure that others build on can scale reach far faster than selling one product. ▸ Watch the gap between reach and revenue: an infrastructure pivot must ultimately monetise. 10. Neko Health — brand, hardware and preventive medicine HQ / country Stockholm, Sweden Founded 2018 (publicly launched 2023) Founders Hjalmar Nilsonne (CEO), Daniel Ek (chairman) Sub-sector Preventive health / consumer medtech (hardware + AI + clinics) Funding to date ~$325m+ (Series B $260m at ~$1.8bn, 2025) Key backers Lightspeed, General Catalyst, Lakestar, Atomico Valuation ~$1.8bn (January 2025) AI posture Hardware-and-AI-native — proprietary scanner plus AI analysis What it does and why it matters Neko Health operates AI-powered, non-invasive full-body scanning clinics that capture millions of data points, skin and moles, cardiovascular and metabolic markers, in minutes, for early detection and prevention. Co-founded in 2018 by Hjalmar Nilsonne and Spotify’s Daniel Ek, it scanned roughly 10,000 patients across Stockholm and London by early 2025, with a waitlist that grew past 100,000. It matters as a high-conviction bet on shifting healthcare from treatment to prevention, and as a study in how a trusted founder brand and a premium consumer experience can generate demand at remarkable speed. Strategy and tactics The strategy is to make preventive, broad-spectrum screening fast, affordable, convenient and consumer-friendly. Tactics include building proprietary scanning hardware and AI analysis in-house (vertical integration), a premium consumer-pay model with a curated, retail-quality clinic experience, demand generation through waitlists and Ek’s brand halo, deliberate one-market-at-a-time geographic rollout, and reinvestment in diagnostics to widen what the scans can detect. Scalable, sustainable, defendable Scalability comes from a repeatable clinic-plus-standardised-hardware unit model, where each new clinic replicates a proven format. Sustainability is supported by direct consumer revenue (around £299 per scan), reducing reliance on slow public payers, though the model is capital-intensive. Defensibility rests on proprietary scanner hardware and IP, a strong consumer brand, and a fast-growing longitudinal dataset that improves the AI with every scan, a data network effect. AI-native versus AI-enabler Neko is best described as hardware-and-AI-native. It was built around proprietary scanning hardware and AI from day one, with AI interpreting the data. It is not a pure-software AI company, but AI is core to the product, sitting between Cera’s adopter stance and Corti’s pure AI-native model. Risks and critical decisions The defining risk is clinical: preventive whole-body screening attracts overdiagnosis and false-positive criticism from clinicians, and faces variable regulation across markets including the US FDA. The model is also capital-intensive and reliant on out-of-pocket demand. Three critical decisions stand out: building proprietary scanning hardware in-house rather than assembling off-the-shelf devices; choosing a consumer-pay premium model over chasing public payers first; and a deliberate, evidence-led geographic expansion rather than blitz-scaling. Founder lessons from Neko Health ▸ A trusted founder brand and a premium experience can build demand extraordinarily fast in healthcare. ▸ Owning the hardware creates a durable moat and proprietary data. ▸ Preventive screening lives or dies on clinical evidence and avoiding overdiagnosis — earn regulator and clinician trust early. ▸ Consumer-pay can sidestep slow public payers, but it is capital-intensive and demand-sensitive. The Nelson Advisors Guide for Founders to HealthTech and MedTech Success in 2026: 10 European Case Studies Part Two: The patterns that decide success AI-native versus AI enablers: a distinction founders over-weight It has become fashionable to divide health-technology companies into the “AI-native”, built from the ground up around machine learning, and “AI enablers” that adopt AI to improve an existing business. Our ten companies map cleanly onto a spectrum. Owkin and Corti are genuinely AI-native, with foundational models at their core. SOPHiA and Ada are AI-native within their niches, built around proprietary engines, though Ada’s predates the LLM era. Neko is hardware-and-AI-native. Oxford Nanopore is AI-enabled deep tech, with neural-network basecalling on a physics moat. And Doctolib, CMR Surgical and Cera are AI enablers, platforms, robots and care operations with AI layered on top. The striking conclusion is that the AI-native label correlates poorly with durability. Several of the most defendable businesses here, Doctolib, Cera, CMR, are emphatically not AI-native, while at least one cautionary tale in the next chapter was an AI-native company whose AI was over-hyped. What consistently matters is not whether AI is the origin of the company but whether AI sits on top of a genuine, hard-to-copy advantage in data, distribution, regulation or hardware. This has a direct implication for founders. The right question is not “am I an AI company?” but “what is my durable advantage, and does AI compound it?” In every successful case here, the AI is valuable precisely because it is fed by proprietary data or embedded in a workflow the company already owns. Where AI is the only differentiator, it is rarely a moat: in 2026, general-purpose models are cheap, capable and improving fast, and a thin AI wrapper is the most easily disrupted position in the market. How European health technology moats are actually built Across all ten companies, four moat types recur, almost always in combination. Understanding which ones you are building and how long each takes, is central to a defensible strategy. • Proprietary, longitudinal data. The single most common moat. SOPHiA’s two-million-profile dataset, Cera’s daily home-visit data, Corti’s clinical interactions, Owkin’s federated hospital data and Neko’s scan dataset all improve their products with scale and cannot be bought off the shelf. The key insight: data moats are usually a by-product of owning a workflow or delivery channel, not a thing acquired directly. • Regulatory clearance and clinical validation. Ada’s EU-MDR certification, CMR’s FDA De Novo and 510(k) clearances and Oxford Nanopore’s clinical validation are assets, not just compliance. They take years and capital to obtain, which is exactly what makes them defensible against fast followers. • Workflow entrenchment and switching costs. Doctolib as a practice’s system of record, Kry’s employed-clinician network, CMR’s trained surgeons and Cera’s commissioner relationships all make the company painful to rip out once embedded. In health, where safety and continuity matter, switching costs are unusually high. • Intellectual property and deep science. Oxford Nanopore’s litigated patent estate and CMR’s modular-architecture IP protect genuinely hard engineering. This moat is strongest where the underlying science is difficult and patient capital has funded a long lead. The pattern is that the strongest companies stack several of these. Doctolib combines network effects, switching costs, data and brand; Oxford Nanopore combines IP, razor-and-blade economics and an ecosystem of published users. A single moat can be eroded; a stack of complementary moats is what produces durability. Founders should map their intended moats explicitly and ask, for each, how many years and how much capital a well-funded competitor would need to replicate it. The funding map: grants, venture, private equity and public markets The capital journeys of these ten companies reveal as much about strategy as their products do. Several lessons stand out for founders planning their own financing. Grants and non-dilutive capital play a real but limited role. Oxford Nanopore benefited from early NIH and Innovate UK support and Owkin from French sovereign backing through Bpifrance, but in every case grants were catalytic seed money, not the engine. For deep-science companies they de-risk the earliest, least investable phase; founders should pursue them aggressively but not mistake them for a business model. Venture capital remains the dominant fuel, and the rounds here are large: Doctolib’s €500m Series F, CMR’s $600m Series D led by SoftBank, Kry’s ~$300m Series D, Neko’s $260m Series B. The clear pattern is that hardware and deep science (Oxford Nanopore raised over $1.1bn before listing; CMR around $1.4–$1.5bn) require far more capital and patience than software. Founders in capital-intensive categories must plan for a decade-long financing arc and choose investors who can sustain it. Strategic and quasi-private-equity capital is a distinctive European feature. Owkin’s decision to sell equity to Sanofi and Bristol Myers Squibb, the customers whose problems it solves, brought data, credibility and distribution alongside money, at the cost of dependency. Eurazeo’s lead of Doctolib’s 2022 round and pension funds such as Ontario Teachers’ and CPP Investments backing Kry show how growth and PE-style capital now anchor the late-stage European market. The public markets are the hardest lesson. Of the two companies here that listed, Oxford Nanopore and SOPHiA, both saw their shares fall roughly 70–75% from peak despite sound underlying businesses. The message is not that listing is wrong, but that an IPO is the beginning of public-market discipline and a milestone valuation must then be grown into, not celebrated as an exit. Risks and rewards: lessons from the casualties Every company in this report carries real risk, regulatory scrutiny, payer dependence, capital intensity, competition from better-funded incumbents and, for the AI-natives, the relentless advance of general-purpose models. But the sharpest lessons come from the companies that failed, and three cautionary cases should be studied by every founder. • Babylon Health. The headline collapse. An AI symptom-checker and telehealth company valued at $4.2bn at its 2021 SPAC listing, bankrupt and sold for parts by 2023. It over-hyped AI that was reportedly closer to a decision tree, took on a ruinous US value-based-care model that paid out almost as much in medical claims as it earned, and expanded recklessly across the UK, US and Africa before reaching profitability. • Pear Therapeutics. The first FDA-cleared prescription digital therapeutics company, bankrupt in 2023 with assets auctioned for a few million dollars. The lesson is brutal: regulatory clearance does not equal reimbursement. Payers would not pay, so there was no viable revenue model despite a validated product. • Olive AI. A healthcare-automation unicorn that raised around $400m at a $4bn valuation and wound down in 2023. It over-promised on AI-driven savings, expanded too fast and failed to deliver — destroying trust and capital. The common thread is unmistakable: inflated AI claims without clinical evidence, ignoring the reimbursement and payer reality, and scaling before proving unit economics. These are precisely the traps the survivors in this report avoided. Kry pivoted to discipline where Babylon chased growth. Cera and Ada reached profitability where Olive burned cash. Neko is deliberately building clinical evidence to avoid the over diagnosis critique. The reward for getting this right is enormous, the companies here have created billions in value and, in several cases, genuine improvements in care, but the asymmetry is severe, and the graveyard is full of well-funded companies that confused a large round for a durable business. The critical decisions that recur Reading across thirty-odd critical decisions made by these founders, a handful recur often enough to count as patterns worth internalising. • Choosing distribution as a moat. Doctolib’s human field sales and Kry’s employed clinicians were expensive, unfashionable choices that became the moat. • Sequencing markets deliberately. CMR proved Versius internationally before the US; Neko expanded one city at a time; Kry localised rather than copy-pasting. • Owning the hard part. Cera owns care delivery, Neko owns the hardware, Owkin owns the data access, each chose the difficult, defensible layer over the easy, copyable one. • Pivoting decisively under pressure. Ada and Kry pivoted to profitability in the 2022 downturn; Owkin and Corti pivoted from services to scalable platforms; CMR delayed a launch for the right product. • Treating regulation and IP as strategy. Ada’s certification, CMR’s clearances and Oxford Nanopore’s re-engineering around a patent suit were all offensive moves, not defensive chores. Commercial models: who pays and how you reach them If the moat determines whether a company survives, the commercial model determines how fast it grows and how much capital it consumes along the way. The ten companies here use markedly different routes to revenue, and the choice of model is one of the most consequential a founder makes. Three broad commercial archetypes appear. The first is recurring software subscription, used by Doctolib and Ada Health, where practitioners or enterprises pay a monthly or annual fee. This model carries the highest gross margins and the cleanest scalability, but it requires solving the cold-start problem of adoption in a conservative profession, which is why Doctolib invested so heavily in human sales. The second is the razor-and-blade hardware model, used by Oxford Nanopore and CMR Surgical, where an installed base of instruments drives recurring consumable and service revenue. This model produces durable, sticky revenue but demands enormous upfront capital and long sales cycles. The third is service or care-delivery revenue, used by Cera and Kry, where the company is paid per episode of care, usually by a public payer; margins are thinner and the business is operationally heavy, but the revenue is recurring, defensive and anchored to genuine demand. Cutting across these is the question of who pays. In Europe the dominant payer is the state, national health systems, regional authorities and statutory insurers, which shapes everything. Public payers are slow to procure and price-sensitive, but once won they provide stable, large-volume, recurring demand and a powerful reference. Cera’s cost-saving argument to the NHS, Kry’s roughly 40% cost advantage over in-person care, and Doctolib’s role in national vaccination booking all turned public-payer alignment into a growth engine. The contrasting route is consumer-pay, chosen by Neko Health, which sidesteps slow procurement entirely but trades it for the need to generate consumer demand and the capital intensity of a clinic network. Owkin and Corti, meanwhile, sell to enterprises, pharmaceutical companies and health systems, where contracts are large but sales cycles are long and concentration risk is real. The lesson for founders is to choose the commercial model that matches both the product and the appetite of available capital, and then to design the company around its payer from day one. The companies that struggled were frequently those that built a product first and discovered the payer later, Pear Therapeutics being the starkest example, with a cleared product and no one willing to pay for it. The European advantage: constraints that became strengths It is tempting to view Europe’s fragmentation, regulatory density and conservative payers purely as handicaps relative to the larger, more homogeneous US market. The companies in this report suggest a more nuanced picture: several of the continent’s apparent disadvantages, handled well, became sources of durable competitive strength. Fragmentation forced discipline. Because there is no single European market, companies such as Kry, Doctolib and CMR Surgical had to master genuine localisation, different languages, reimbursement systems, regulators and clinical cultures. That is harder and slower than scaling across fifty US states, but the resulting multi-country regulatory and operational know-how is itself a moat that new entrants cannot quickly replicate. The very friction that slowed these companies down also protected them once they had crossed it. Strong public health systems created uniquely valuable data and demand. Europe’s universal, largely single-payer systems generate longitudinal, population-scale clinical data and a coherent buyer with a direct interest in cost reduction. Owkin’s access to leading oncology hospitals, Cera’s entrenchment with NHS commissioners, SOPHiA’s network of clinical institutions and Corti’s reach into the NHS were all enabled by the structure of European healthcare. The same systems that are slow to buy are, once aligned, extraordinary sources of data and stable demand. World-class science and a deepening talent base supplied the raw material. Oxford Nanopore emerged from the University of Oxford, SOPHiA from EPFL, Owkin from the French AI research ecosystem and Corti from Danish machine-learning research. Europe’s universities and research institutions remain a genuine comparative advantage in deep-science health technology, and a maturing pool of operators, many of whom cut their teeth at the first wave of European technology champions, increasingly supplies the commercial talent that was historically the weak link. None of this makes Europe an easy place to build. Capital is still thinner at the latest stages, exits are harder, and the public-market reception for Oxford Nanopore and SOPHiA was punishing. But the founders who treated Europe’s constraints as design parameters rather than obstacles built companies that are, in several respects, more defensible than a faster-moving US equivalent would have been. Timing, sequencing and the discipline of focus A final pattern deserves its own treatment because it appears, in some form, in almost every successful case and almost every failure: the discipline of sequencing, doing the right things in the right order and the courage to stay focused under the pressure to expand. The casualties shared a failure of sequencing. Babylon Health scaled across three continents and into a complex US value-based-care model before it had proven profitability anywhere; Olive AI broadened its product line before it had reliably delivered the savings it promised on a narrower one. The survivors did the opposite. CMR Surgical refused to enter the US until it had a cleared, competitive product, even scrapping a planned launch to wait. Neko Health expanded one city at a time, building clinical evidence as it went. Kry cut roughly four hundred jobs and exited Germany in 2022 to concentrate on markets it could make profitable. Each of these was a decision to do less, sooner, in order to do more, later. Timing also mattered enormously at the level of capital. The companies that raised large rounds at the 2021 peak and then spent as though the environment would persist, Owkin has been admirably candid about over-hiring, had to retrench painfully. Those that treated the downturn as a forcing function for profitability emerged stronger and, in Cera and Ada’s cases, were able to raise again from a position of strength. For founders, the practical implication is to raise when you can but spend as though you cannot, and to treat every expansion, new geography, new product line, new payer type, as a decision that must be earned by proof in the current one, not assumed by ambition. Focus, in this market, is not a constraint on growth; it is the mechanism of durable growth. The companies that endured were those willing to be smaller than their funding allowed until each step was proven, and the discipline to sequence carefully is perhaps the single most transferable lesson in this guide. Conclusion: a founder’s checklist for 2026 The ten companies in this guide are different in almost every respect, software and hardware, consumer and enterprise, profitable and pre-profit, AI-native and AI-enabling, yet they converge on a small set of principles that should shape any founder’s strategy in 2026. Build your business around a durable advantage, not a feature. In European health technology that advantage is almost always proprietary data, regulatory clearance, workflow entrenchment, deep-science IP or some stack of these. Ask continually how many years and how much capital a well-funded rival would need to copy what you have built. Be honest about what kind of company you are. If you are capital-intensive deep science or hardware, plan for a decade-long financing arc and choose patient, deep-pocketed investors. If you are software, the prize is scale and margin, but the AI layer alone will not protect you. Either way, get close to whoever actually pays for healthcare, in Europe that is usually a public payer or a clinician and prove the unit economics before you scale. Treat AI as a compounding layer, not an identity. The most defendable companies here use AI to amplify a moat they already own. A thin AI wrapper on someone else’s data or workflow is the most easily disrupted position in the market. Respect the asymmetry. The rewards in this market are extraordinary, but the failures, Babylon, Pear, Olive, were not small companies; they were well-funded, celebrated businesses that mistook a large valuation for a durable one. Discipline, evidence and proximity to the payer are what separated the survivors from the casualties. They will do so again in 2026. Looking ahead, the next wave of European health-technology value is likely to accrue to companies that combine a proprietary clinical-data asset with the new generation of AI, not as a thin application layer, but embedded in workflows and regulatory pathways that took years to earn. The maturation of generative AI lowers the cost of building a feature but raises the premium on the things AI cannot easily manufacture: trusted data, clinical evidence, regulatory clearance and entrenchment with the people who deliver and pay for care. The founders who understand that distinction, who use 2026’s cheap intelligence to compound an advantage rather than to substitute for one, will build the defendable companies of the coming decade. The ten companies in this guide have already shown how it is done. None had an easy path; several are still on theirs. But each, in its own way, made the hard, defensible, sometimes unfashionable choices that the European health-technology market rewards over time. For a founder setting out in 2026, that is the most valuable inheritance of all: not a formula, but a set of proven instincts about where durable value in this industry truly comes from. The Nelson Advisors founder checklist ▸ Name your moat(s) explicitly — data, regulation, workflow, IP — and quantify how long they take to replicate. ▸ Match your financing plan to your capital intensity; pursue non-dilutive grants early but never mistake them for a model. ▸ Get to the payer and prove unit economics before scaling; reimbursement is as important as regulatory clearance. ▸ Use AI to compound an existing advantage; avoid being a thin wrapper on someone else’s data. ▸ In a downturn, choose profitability and discipline over momentum — it is what kept the survivors alive. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

Scaling Healthcare Innovation: Life Sciences, MedTech and HealthTech Success Stories from the Goldman Sachs 10,000 Small Businesses UK Programme The persistent productivity puzzle within the United Kingdom remains a major structural challenge for economists, academic institutions, and policymakers. While the British economy excels at spawning startups, transition rates from early-stage micro-enterprises to scaling, mid-market companies have historically been constrained by a pronounced management and leadership support gap. This barrier is particularly challenging in highly technical, regulated, and capital-intensive sectors such as life sciences, biotechnology, medtech, and healthtech, where long research and development cycles, complex clinical validation pathways, and talent acquisition hurdles compound the difficulties of scaling. To bridge this critical growth deficit, the Goldman Sachs 10,000 Small Businesses (10KSB) UK programme has acted as a vital educational and strategic intervention since its launch in 2010. Fully funded by the Goldman Sachs Foundation, the initiative provides high-growth small business leaders and social entrepreneurs with 100 hours of practical business and management education. Developed by leading experts, the curriculum is delivered in partnership with top-tier academic institutions, including Saïd Business School at the University of Oxford, the Aston Centre for Growth at Aston University, Leeds University Business School at the University of Leeds, and the Manchester Metropolitan University Business School. By targeting decision-makers of companies operating for at least three years with 5 to 50 employees and annual revenues exceeding £250,000, the programme provides the structural tools necessary to design and execute a robust, customized Business Growth Plan. An analysis of health-related business success stories within this framework shows how targeted business education helps scientific and clinical founders scale their companies, navigate market challenges, and create high-value employment. Longitudinal Economic Impact and the 10KSB Programme Effect The macroeconomic contribution of the 10KSB UK programme has expanded considerably over its fifteen-year history. Evaluations conducted by the Enterprise Research Centre (ERC) demonstrate that programme graduates achieve significantly higher rates of job creation, revenue generation, and productivity growth compared to matched control groups of similar size, age, and sector. The growth trajectory of the programme's aggregate economic footprint is detailed in the table below: Longitudinal Milestones (Year of Report) Total Scaling Graduates Combined Annual Revenue Estimated Combined Headcount Key Operational & Strategic Indicators Early Evaluation Phase (2015) 2,300+ (US & UK combined sites) N/A (Initial US/UK baseline collection) N/A (Baseline operational metrics) 66.8% of graduates reported revenue growth within 6 months; 76.0% grew revenues within 18 months. Middle Scaling Phase (2022) 1,900+ (UK graduates) £2.8 billion N/A (Focus on pandemic retention) 63% invested in R&D during the pandemic; 62% planned continued R&D investment over 2 years. Advanced Scaling Phase (2024) 2,300+ (UK graduates) £5.9 billion 53,000+ employees 47% expanded workforce during macro headwinds; 26% recorded annual turnover growth above 20%. 15-Year Milestone (2025) 2,500+ (UK graduates) £10.6 billion 82,000+ employees Combined 15-year net addition of £2.7 billion in revenue and 41,313 jobs directly attributable to the programme. This expansion is supported by deep behavioural changes. Following graduation, 94% of participants report higher confidence in managing scaling challenges, 84% systematically utilise financial data to drive strategic decisions, 97% feel more effective as business leaders, and 90% introduce new corporate processes or organisational systems. According to the Enterprise Research Centre, three years after completing the programme, 10KSB UK graduates outperform similar companies by increasing revenues by up to 43% more and growing their staff by over a third. The programme's impact also includes a 14% outperformance in productivity compared to equivalent companies, directly addressing the UK's broader productivity challenges. Pandemic Adaptation and Resilience Strategies The structural resilience of 10KSB UK alumni was tested during the COVID-19 pandemic. While SMEs generally faced severe disruptions, 10KSB graduates pivoted quickly by using the strategic frameworks of their business education. To support its network, the 10KSB UK programme launched the "Reimagining Business" initiative between March and May 2020, delivering targeted webinars, weekly coaching sessions, and faculty insights to over 450 small business leaders. This continuous support helped guide major pivots. Approximately 80% of programme participants revised their pre-pandemic business models, 68% adapted their operational structures, 69% successfully launched new products or services, and 88% accelerated their adoption of digital technologies. Within health-related sectors, many women-run and founder-led businesses exposed to healthcare, technology, or remote-capable operations reported revenue increases during the pandemic by adapting to meet shifting clinical and clinical-support needs. Sectoral Success Stories in the Life Sciences, Medtech and Healthtech The practical application of the 10KSB curriculum is best demonstrated through the growth of specific companies in the life sciences, medtech, and healthtech fields. These firms have successfully converted business training into scalable clinical and commercial operations. CatSci: Transitioning from Catalyst Screening to Global Drug Development CatSci Ltd, based in Cardiff, Wales, is a prominent life sciences success story that illustrates the journey from a specialised laboratory service to a globally recognised pharmaceutical innovation partner. Founded in 2011 by scientists following a spin-out from AstraZeneca, the business was initially established as a catalyst screening company. By 2019, CatSci operated with a turnover of £3 Million and employed 25 people. To scale further, Chief Executive Officer Dr. Ross Burn completed the 10KSB UK programme at Oxford Saïd Business School to build the commercial systems needed to expand the firm's global presence. Through the programme, Burn refined CatSci’s strategy and transitioned the company's focus to the broader, high-value field of process research and development for new medicines, with a particular focus on the development of scalable oligonucleotide therapeutics to support emerging RNA-based pipelines. This repositioning helped the firm grow past its initial targets. Between 2018 and 2022, CatSci grew its revenues by an average of 50% annually, expanded its headcount to 76, and grew its international trade footprint across 13 countries—including entry into the Japanese, European, and US markets. This growth in international trade earned CatSci the Queen’s Award for Enterprise in International Trade in 2022. By building solid financial and operational structures, CatSci successfully positioned itself to secure a major institutional investment from European private equity firm Keensight Capital, helping fund its continued laboratory expansion. https://catsci.com/about-us/our-history/ June Medical: Scaling Medtech Manufacturing through Operational Excellence June Medical, an award-winning medical device manufacturer founded by Angela Spång, shows how medtech scale-ups can use process innovation and structured workforce development to capture international market share. June Medical is best known for developing and manufacturing single-use, plastic injection molded surgical retraction systems made of advanced Terlux polymers, which improve surgical access and sterile control in urological procedures. Spång enrolled in the 10KSB UK programme, a curriculum she compared to a condensed Executive MBA, to build the management capacity required to scale her proprietary medical products globally. The programme challenged her to redefine her company's value proposition and transition from a localised distributor to an international IP-led manufacturer. This strategic development culminated in June Medical being named the "Most Innovative Business" winner across the entire 10KSB UK alumni community during the programme's 15th-anniversary milestone celebration at the Tate Modern in 2025. To support this rapid product growth, June Medical addressed the technical skills shortage in manufacturing by implementing a structured apprenticeship scheme. By standardising its training and operational processes, the company successfully integrated young apprentices into its commercial workflows, helping expand operations at its UK facilities and launch its US-focused export pipelines. https://junemedical.com Principle Healthcare: Internationalising Nutraceutical Manufacturing Principle Healthcare Group, based in Skipton, North Yorkshire, highlights how life sciences and consumer-health manufacturers can utilise regional business support to scale across Europe. Principle Healthcare is a major producer of vitamins, minerals and dietary supplements, with capabilities spanning formulation, advanced packaging, and large-scale manufacturing. Through its engagement with Leeds University Business School, one of the key 10KSB UK delivery partners, the firm integrated advanced marketing, supply-chain logistics, and export strategies into its core operations. These improvements allowed Principle Healthcare to expand its international trade footprint, leading to the company winning the UKTI Export Achievement Award at the Medilink Yorkshire and Humber Awards. The Medilink Awards celebrate the commercial achievements of healthcare technology and life sciences organisations in the region. By building structured commercial partnerships, Principle Healthcare scaled its international brand presence and secured a market-leading position. This successful trajectory made it an attractive strategic partner, culminating in its acquisition by the global healthcare group, helping cement its position in the European nutraceutical sector. https://www.principlehealthcare.com Concierge Medical: Systematising Rural Private Healthcare Delivery Concierge Medical, founded in 2013, operates a private, home-visiting general practice that provides continuous primary care across the Cotswolds and surrounding rural areas. The business model addresses a geographical challenge where access to traditional, physical primary care facilities can be difficult. Kat Carrick, representing Concierge Medical, completed the 10KSB UK curriculum to address operational constraints that were bottlenecking the company's growth. The clinical founders needed structured management frameworks to transition from a localised, hands-on medical practice into a scalable, membership-based healthcare business. The programme provided the tools necessary to analyse and resolve operational inefficiencies, leading to the design of a standardised, technology-enabled primary care platform. Following graduation, Concierge Medical recorded a 98% increase in its subscription membership base within six months. The confidence and operational discipline gained during the course helped the leadership set more ambitious growth targets, prompting the company to actively seek institutional external investment to scale its rural concierge model to other underserved regions across the UK. https://www.conciergemedical.co.uk/about-us/ Myonex: Rebranding and Scaling Global Clinical Trial Supply Channels Myonex, formerly known as Myoderm, shows how clinical trial support services can pivot their brand and operational footprint to capture global pharmaceutical demand. While the company had built a strong reputation in clinical trial supply within the United States, it initially struggled to scale its brand internationally. A major barrier was a widespread misconception among global audiences that the suffix "-derm" in the company's original name limited its services to dermatology clinical trials. To support its global expansion and clear up this market misconception, the company underwent a comprehensive rebranding process, officially launching as Myonex in 2020. This rebrand was supported by extensive market research, including interviews and surveys with life sciences professionals across Europe and North America. The strategic shift helped accelerate growth, with Myonex recording annual growth rates exceeding 20% for three consecutive years. During this scaling phase, the company tripled its global headcount, introduced new services tailored for decentralised clinical trials, opened a new 65,000-square-foot global headquarters in the US, and significantly expanded the clinical supply capacity of its UK-based facilities. https://www.myonex.com/ Ada Jabaru and Nistad Limited: Healthcare Compliance, Care Delivery and Community Advocacy Ada Jabaru, the founder and director of Nistad Limited, represents a unique 10KSB success story at the intersection of regulatory compliance, care delivery, and technology management. With over 17 years of experience in regulatory compliance, risk assessment, and financial audit, Jabaru initially built Nistad Limited to provide compliance solutions for financial institutions working with complex investment and asset management software. However, as a 10KSB UK alumna recognised within the healthcare and medical-related business categories, Jabaru leveraged her expertise in audit and compliance to expand into adult social care and community health systems. Her clinical and care-related portfolio included directing Flourish Adult Care Limited, alongside her leadership of the Nistad Empowerment Foundation. This foundation supports primary healthcare delivery and educational outreach in underserved communities, with a strong focus on training young people and women from disadvantaged backgrounds in AI and digital skills. By using the strategic insights gained from the 10KSB programme, Jabaru has become a prominent voice for SME leadership. She has participated in policy panels with senior Goldman Sachs executives at Mansion House, advocating for increased investment in technology education and digital skills to help solve the UK's skills crisis and improve healthcare access. https://nistadfs.com Scaling Healthcare Innovation: Life Sciences, MedTech and HealthTech Success Stories from the Goldman Sachs 10,000 Small Businesses UK Programme AI-Ready Transformation and Digital Telehealth Ecosystems The convergence of artificial intelligence, real-time analytics, and clinical operations represents the next major growth frontier for healthtech and medtech scale-ups. The 10KSB UK programme has continued to adapt its curriculum to help founders manage this technological transition, focusing on data governance, predictive modelling, and AI-driven clinical workflow automation. A key example of this focus is Data Inc, which was selected for the 10KSB UK programme in 2026. Data Inc specializes in helping scaling businesses transition toward value-assured, AI-ready operating models. For modern healthtech and clinical companies, data democratisation and predictive analytics are essential to maintain data reliability, comply with medical regulations, and protect patient information. By helping founders integrate machine learning and automated quality control, the 10KSB framework enables scaling healthcare companies to deploy predictive tools that improve patient diagnostics and operational agility. This systemic integration of AI within telehealth ecosystems is also illustrated by Ufonia, a healthtech venture supported by the Oxford Foundry, which is closely integrated with the Saïd Business School 10KSB partnership. During the pandemic, Ufonia pivoted its AI-telemedicine engine to address the clinical backlogs caused by cancelled procedures. By automating post-surgery clinical reviews and patient follow-ups via its AI-driven voice platform, Ufonia helped hospitals streamline post-operative workflows, demonstrate the efficacy of automated care, and reduce administrative burdens on frontline healthcare workers. Overcoming Structural Barriers to Scale in High-Growth Health Sectors An analysis of the success stories within the 10KSB UK alumni network reveals how specific scaling barriers in the healthcare and life sciences sectors can be systematically addressed through targeted business education: Distinct Scaling Challenge in Health Sectors Causal Impact on Business Growth 10KSB Educational & Strategic Solution Observed Operational Outcome The Technical Founder Dilemma Scientific or clinical founders often struggle with the transition from technical experts to strategic CEOs, leading to operational bottlenecks. Modules on Leadership & Culture focus on strategic delegation, process standardisation, and defining a clear corporate mission. Clinical and scientific founders successfully transition to strategic business development, enabling larger-scale operations. The Technical Skills & Talent Crisis 45% of fast-growth SMEs report an inability to access the technical, digital, and data talent needed to sustain their scaling plans. SME Talent Acquisition Strategy helps founders build internal training systems and structured apprenticeship pathways. June Medical implemented an apprenticeship model that brought in and trained young talent, supporting global export growth. Strict Regulatory Compliance & Auditing Medtech and life sciences products face strict regulatory approval and clinical compliance pathways before market entry. Modules on Strategic Operations & Metricsprovide frameworks to build audited socio-technical operating models. Data Inc and Nistad optimized their governance structures, helping de-risk their offerings for institutional partners. High R&D Capital Requirements Long clinical development cycles require significant upfront capital before generating sustainable commercial revenues. Cash & Funding Modulesteach business valuation, pitch readiness, and how to align with patient venture capital. CatSci secured growth funding from Keensight Capital; Concierge Medical positioned its model for institutional investment. Navigating the Founder-to-CEO Transition The transition from a technical founder to a strategic executive is a key tipping point for scaling healthcare businesses. Clinical and scientific founders are trained to focus heavily on scientific accuracy and patient care, which can sometimes lead to a highly centralised management style. The 10KSB curriculum addresses this by helping founders build a structured management team, delegate day-to-day operations, and focus on long-term corporate positioning. This shift is clear in the case of CatSci, where the transition from a localised laboratory provider to an international therapeutic development partner was enabled by establishing clear operational roles and institutional governance. Standardising Operations to Enable Scalable Care Subscription-based primary care and medical services are structurally limited by the physical capacity of their clinical providers. To scale, these models must standardize their administrative and clinical workflows. The 10KSB UK programme’s focus on strategic operations helps healthcare businesses utilise technology to automate scheduling, billing, and patient follow-ups. This process optimisation allowed Concierge Medical to grow its subscriber base by 98% in the months following graduation without needing to add proportional administrative staff, showing how standardising operations can help expand clinical access. Policy Advocacy, Generation Growth and Future Economic Trajectories Beyond direct business support, the 10KSB UK programme has become a significant advocate for small business growth, culminating in the launch of "Generation Growth: The Small Business Manifesto" and "The Growth Agenda: Growth Shots" in 2024 and 2025. These reports, based on detailed surveys of over 550 high-growth 10KSB UK alumni, outline the key policy measures necessary to unlock the full potential of scaling SMEs. The findings show that while access to capital remains a challenge, with 37% of surveyed small business owners reporting difficulties in obtaining the growth finance they need, the skills shortage is the single largest barrier to scaling. The Manifesto urges policymakers to prioritise the up-skilling of the UK workforce, ahead of traditional policy focuses like international trade support and business rate reform. Industry leaders, including ScaleUp Institute CEO Irene Graham OBE and British Business Bank Chair Stephen Welton CBE, have highlighted the importance of these recommendations, emphasising that supporting patient capital and vocational education is vital to unlocking the UK's broader innovation economy. The success stories of CatSci, June Medical, Principle Healthcare, Concierge Medical, Myonex, and Nistad demonstrate that when scientific and clinical expertise is paired with structured management education, high-growth healthcare companies can scale effectively. By standardising internal operations, adopting digital and AI-driven workflows, and building structured talent development pipelines, these businesses successfully navigate strict regulatory environments and create high-value employment. Expanding this educational support across the UK’s broader life sciences, medtech, and healthtech sectors will be key to addressing the nation’s persistent productivity puzzle and driving sustainable, innovation-led economic growth. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

Beyond the Prescription Pad: The Strategic Convergence of Consumer Technology, Retail Infrastructure, and Preventive Healthcare The Macroeconomic Catalyst: The Crisis of Legacy Reactive Care and the Rise of Proactive Healthspan The global healthcare ecosystem is undergoing a profound structural transition. For decades, clinical medicine operated under a reactive, "one-size-fits-all" framework. Under this legacy model, clinical interventions were typically initiated only after the manifestation of symptomatic disease, overlooking the granular genetic, metabolic and physiological variations that define individual baselines. Today, a combination of macroeconomic pressures, demographic shifts and rapid technological acceleration is driving a transition toward active, human-centric health management. This evolution redefines the clinical focus from acute disease suppression to proactive healthspan extension. At the center of this transition is the integration of lifestyle medicine, digital health technologies, and functional diagnostics that look past the traditional prescription pad. Traditional clinical systems are increasingly burdened by aging populations, flat reimbursement rates, severe healthcare worker shortages, and rising operational overhead. These systemic vulnerabilities have compressed patient-provider interactions, leaving primary care physicians with insufficient time to address the root lifestyle behaviours such as nutrition, chronic stress, sleep architecture and metabolic fitness, that drive over 80% of chronic disease burdens. The scale of this primary care vacuum is significant; the Association of American Medical Colleges (AAMC) predicts a shortage of up to 55,000 primary care physicians in the United States over the next decade. This structural shortage has accelerated the growth of the global smart healthcare industry, which expanded from an estimated market size of $147.7 Billion in 2019 to $181.7 Billion in 2022, maintaining an average annual growth rate of 11.0%. As these platforms mature, healthcare delivery is shifting from episodic, product-centric transactions to integrated, platform-based solutions. This commercial evolution has created a bifurcated wellness and healthcare market. In the United States, fragmented insurance models and high operational friction have paved the way for a direct-access, cash-pay longevity ecosystem. Conversely, European markets view public healthcare as a social safety net, making European consumers less likely to pay directly for basic care. Startups operating in Europe must therefore package preventive health services with social capital, lifestyle experiences and identity markers to persuade consumers to pay out-of-pocket for services they expect to receive for free. Market Dimension Smart Healthcare Market (2019) Smart Healthcare Market (2022) Project CAGR Global Valuation $147.7 Billion $181.7 Billion 11.0% These regional market dynamics reflect broader socio-economic disparities. Sociological research indicates that health vulnerabilities scale with economic deprivation. For example, individuals in the most deprived deciles are twice as likely to live with five or more chronic conditions by late life compared to those in the least deprived deciles. Furthermore, individuals in high-resource areas are roughly 40% less likely to progress from a single chronic condition to multi-morbidity. To address these deep-seated disparities, frameworks such as the United Kingdom's "social prescribing" model connect older adults with non-pharmacological interventions like local walking groups, bereavement services, and community kitchens. This model addresses the social determinants of health directly, recognizing that loneliness and depression are primary drivers of clinical non-adherence. Proactive social connections have been shown to reverse cognitive and physiological decline, demonstrating that clinical outcomes are deeply connected to a patient's broader social environment. The integration of lifestyle medicine also extends to evidence-based complementary modalities funded by research bodies like the National Institutes of Health (NIH) National Center for Complementary and Integrative Health. Evidence-based therapies such as yoga and acupuncture, which utilises fine filiform needles across approximately 360 to 380 specific points in the human body, are increasingly used to manage chronic lower back pain, anxiety and depression while reducing dependency on prescription medications. Clinical observations indicate that incorporating weekly acupuncture sessions can enable chronic pain patients to reduce high-dose opioid and narcotic regimens by half within three months, and eliminate usage entirely by six months. By focusing on whole, plant-based foods rather than expensive synthetic supplements, functional medicine practitioners are demonstrating that dietary modifications can prevent clinical deterioration and reduce overall healthcare spending. Health Management Modality Target Biomarker / Symptom Clinical Evidence & Efficacy Weekly Acupuncture Chronic pain & opioid dependency. 50% dosage reduction at 3 months; complete cessation at 6 months. Yoga & Mindfulness Lower back pain, anxiety, & depression. Validated improvement in flexibility, balance, and mental health markers. "Food as Medicine" (FIM) Metabolic syndrome & diabetes risk. Whole-food, plant-based diets reduce cardiovascular risk by 80% and diabetes risk by 90%. Social Prescribing Geriatric multi-morbidity & isolation. Reverses cognitive decline and improves medication adherence. Passive Continuous Phenotyping and Remote Therapeutic Monitoring: The ŌURA Integration Framework The transition away from episodic clinical care requires a shift toward continuous passive digital phenotyping. While traditional medical records capture only static, point-in-time data during annual physicals, consumer wearables gather continuous physiological metrics in the user’s natural environment. This real-world evidence establishes highly personalised baselines, allowing algorithms to detect micro-deviations before they manifest as acute clinical events. Within this wearable landscape, the ŌURA Ring has transitioned from a consumer wellness tracker into a clinically validated, research-grade remote monitoring platform. The form factor of a smart ring offers distinct advantages over wrist-worn alternatives. By measuring biometrics from the digital arteries of the finger, where the pulse signal is 50 to 100 times stronger than on top of the wrist, the smart ring achieves superior signal-to-noise ratios. This sensor stability translates into excellent data fidelity, particularly for sleep stage classification, heart rate variability (HRV), and resting heart rate (RHR). Independent clinical validation has established the device's accuracy compared to laboratory polysomnography (PSG), the gold standard for sleep evaluation. Research shows that the ŌURA Ring Generation 3 demonstrates approximately 94.4% sensitivity and 91.7% overall sleep accuracy, achieving 79% agreement with PSG in four-stage sleep classification. This level of agreement is remarkably close to the 83% consensus rate typically observed between trained sleep technicians scoring the same recording manually. Furthermore, a 2025 independent study found that the ring exhibited the strongest agreement among consumer wearables for both HRV and RHR measurements. To integrate these capabilities into traditional clinical workflows, ŌURA has moved aggressively to establish direct APIs with electronic health record (EHR) systems and clinical platforms. These integrations allow consumer-generated data to flow passively into platforms like Epic via consumer portals such as MyChart, eliminating patient data entry burdens. Rather than presenting physicians with raw, uncontextualized metrics, integration software normalises the data, organising it into trends displayed alongside standard clinical charts. Wearable Performance Metric Clinical Validation vs. Polysomnography (PSG) Source Validation Study Sleep Detection Sensitivity 94.4% Sensitivity Svensson et al. 2024 Overall Sleep Accuracy 91.7% Accuracy Svensson et al. 2024 Four-Stage Sleep Classification 79.0% Agreement (PSG scoring baseline is 83.0%) Robbins et al. 2025 Device Wear Adherence 95.0% compliance (24-hour wear time over 8 weeks) Independent Wearable Compliance Study These continuous physiological metrics are particularly valuable for Remote Therapeutic Monitoring (RTM) programs. Under US Medicare guidelines, clinicians can use CPT codes 98975 through 98981 to bill for monitoring non-physiological data such as therapy adherence and musculoskeletal or respiratory response. By combining continuous, passive wearable data with subjective patient-reported outcomes, clinicians can build comprehensive, longitudinal datasets while avoiding the survey fatigue that often undermines traditional remote patient monitoring. Furthermore, because ŌURA’s finger-worn form factor maintains a 95% compliance rate over an eight-week period, outperforming wrist-worn devices that suffer from frequent drop-offs, clinicians can capture more ecologically valid biometric data during everyday routines. This continuous data pipeline has enabled a series of strategic integrations across clinical and research fields: Medicare Advantage Programs: In June 2026, Essence Healthcare integrated ŌURA Ring data into its physician-led clinical care workflows, allowing providers to monitor sleep patterns, recovery metrics, and early markers of illness in Medicare Advantage populations. Digital Women's Health: In May 2026, Twentyeight Health integrated the ring's high-fidelity skin temperature sensors to track reproductive cycles. This integration builds upon the ring's compatibility with FDA-cleared birth control platforms like Natural Cycles, which analyzes temperature trends to determine fertile windows. Academic Research Collaborations: The ring is used in major research initiatives, including the University of California, San Francisco (UCSF) TemPredict project, which tracks early infection patterns, and a University of Vermont study focused on tracking objective sleep and physical recovery metrics to evaluate student mental health. The Retail Retreat and Re-alignment: Deconstructing Walmart's Pivot to Decentralised Health Operations The physical footprint of major retailers has long been viewed as a potential solution to the accessibility and affordability issues of the American healthcare system. However, the operational realities of brick-and-mortar primary care delivery have forced a major strategic re-evaluation. In 2024, retail health faced significant disruption. Industry giants like Walgreens closed approximately half of their VillageMD clinics, CVS Health realigned its leadership amid cost pressures within its Aetna Medicare Advantage segment, and Walmart announced the complete closure of all 51 in-store health centres and the termination of its Walmart Health Virtual Care platform (formerly MeMD). The closure of Walmart Health centers after a five-year effort highlights the difficulties of operating physical clinics under traditional fee-for-service reimbursement models. Positioned to serve low-income and medically underserved communities, Walmart’s clinics struggled with flat insurance reimbursement rates, rising labor costs, and a nationwide shortage of healthcare workers. Unlike selling consumer goods, clinical healthcare delivery requires specialised, highly compensated personnel, complex billing infrastructures, and extensive liability management, factors that clashed with Walmart's high-volume, low-margin retail business model. Retail Health Entity 2024–2026 Operational Realignment Details Associated Labour & Footprint Reductions Walmart Health Closed all 51 physical health centers and terminated its virtual care platform. 90 days of transition pay for clinical associates and providers. Walgreens (VillageMD) Closed 160 VillageMD clinics, shifting away from an exclusive reliance on a single-provider model. Shuttering 1,200 retail locations (14% of footprint) over three years. CVS Health (Aetna) Replaced CEO Karen Lynch with David Joyner; faced Medicare Advantage margin pressures. Laid off 2,900 employees, closed 29 retail pharmacies, and cut Coram infusion services. Amazon Health Services Repositioned digital assets to focus on virtual care and Amazon Pharmacy. Eliminated several hundred roles across One Medical and Amazon Pharmacy. This retail retreat has shifted the competitive landscape. While CVS has maintained its commitment to physical clinics by operating over 200 stand-alone Oak Street Health centers across 25 states, other retailers are shifting to asset-light, digital-first models. Rather than managing clinical clinics directly, Walmart is leveraging its core strengths: its retail data, its fresh food supply chain, and its network of 4,600 physical pharmacies. Because over 4,000 of Walmart's stores are located in federally designated medical provider shortage areas, its in-store pharmacies and 3,000 Vision Centers often serve as the primary entry point for local healthcare. This infrastructure allows Walmart to offer essential services like immunisations, point-of-care testing, and medication therapy management without the overhead of operating dedicated physical clinics. A key component of this post-clinic strategy is the Walmart Healthcare Research Institute (WHRI), launched in 2022 to connect retail reach with clinical trials. In January 2026, Walmart announced a partnership with clinical research company Care Access to open four clinical research sites in spring 2026, situated at former clinic locations and a rural retail store. This model allows Walmart to offer health screenings and clinical study access to diverse, rural, and underserved populations without the burden of maintaining full-service primary care facilities. Simultaneously, Walmart is positioning its stores to capture the trend of Food as Medicine (FIM), particularly in response to the rise of GLP-1 receptor agonists (such as Ozempic and Wegovy). While some food retailers worry about GLP-1s reducing overall caloric intake, Walmart is using its pharmacy infrastructure as a gateway to drive healthy, nutrient-dense food sales. Because GLP-1 therapy requires long-term lifestyle modification, specifically maintaining lean muscle mass through high-protein diets and nutrient-dense foods, the pharmacy can connect prescriptions directly to targeted retail grocery programs. By combining its retail transaction data with digital wallets and filtered-spend technology (e.g., directing SNAP benefits or employer wellness funds toward medically tailored groceries), Walmart is turning its stores into community wellness centres. Integrated FIM & GLP-1 Retail Models Key Platform Capabilities Target Clinical Population Walmart Pharmacy & SNAP Link Connects GLP-1 prescriptions with filtered-spend grocery and "Great for You" tracking. Medically underserved and rural populations. Instacart Health & Dispatch Health Offers seniors aging in place targeted nutrition programs using virtual storefronts and Care Carts. Geriatric populations with multi-morbidity. Season Health & Employer Plans Integrates digital wallets with medically tailored groceries and customised nutrition coaching. Employees utilising GLP-1s for weight loss. H-E-B Wellness Integration Integrates physical retail grocery stores with registered dieticians and curated menu planning. Regionally concentrated families and chronic disease patients. Beyond the Prescription Pad: The Strategic Convergence of Consumer Technology, Retail Infrastructure, and Preventive Healthcare Virtual Clinics as Holistic Benefit Solutions: The Maven Clinic Case Study While retail giants adjust their physical footprints, virtual clinics are proving that digital-first care can drive clinical outcomes and reduce employer costs. Historically, digital health benefits were highly fragmented, with employers adopting isolated point solutions for fertility, maternity, postpartum depression and menopause. This fragmentation created administrative inefficiencies, low user engagement, and disjointed care experiences. This fragmentation is particularly costly given the high financial and operational impact of reproductive health challenges on employers: High Intervention Costs: An average round of IVF in the United States costs approximately $23,000, and more than a third of patients spend over $50,000 over the course of their fertility journeys. Maternity Care Deserts: More than 50% of counties in the United States lack a single practicing OB-GYN, making geographical access a major barrier. Mental Health Losses: Untreated perinatal mood and anxiety disorders result in an average presenteeism cost of $2,871 per affected employee per year, contributing to an estimated $40,478 in economic losses per person due to associated unemployment. Surgical Delivery Rates: The standard C-section rate in the United States stands at 32%, significantly higher than the World Health Organization’s recommended baseline of 10%. This disparity drives clinical costs, with each surgical delivery costing employers an average of $26,000. Maven Clinic, founded in 2014, pioneered a comprehensive, unified family health benefits platform spanning fertility, maternity, pediatrics and menopause. Maven operates as an integrated virtual clinic and benefits administrator, coordinating care through a network of over 30 provider specialties, including reproductive endocrinologists, OB-GYNs, doulas, midwives, lactation consultants, and mental health professionals. This virtual-first architecture provides 24/7 coverage, which is particularly beneficial for remote and hybrid workers or those living in maternity care deserts; in fact, 60% of Maven appointments occur outside standard office hours. Maven's continuous support model has demonstrated clear, peer-reviewed clinical and financial outcomes, establishing a strong business case for corporate benefits integration: NICU Admission Reductions: Maven’s proactive monitoring and educational interventions contribute to an up to 28% reduction in Neonatal Intensive Care Unit (NICU) admissions. C-Section Rate Reductions: By connecting mothers with doulas and personalised birth planning resources, Maven helps lower C-section rates by up to 15%. Alternative Paths to Conception: Among fertility members, 30% achieve pregnancy naturally without requiring expensive assisted reproductive technologies (ART) like in vitro fertilisation (IVF) or intrauterine irritation (IUI). Retention and Return-to-Work: Over 90% (and up to 94% in certain cohorts) of Maven members return to work post-parental leave, compared to a national average of just 57%, saving employers significant recruiting and onboarding costs. Financial ROI: These clinical improvements generate average direct savings of up to $5,000 per member, translating to a validated 2x to 4x return on investment (ROI) for employer clients. Clinical & Financial Outcome Metric Traditional Legacy Healthcare Baseline Maven Integrated Platform Performance NICU Admission Rate Standard industry benchmarks. Up to 28.0% Relative Reduction C-Section Delivery Rate 32.0% US National Average. Up to 15.0% Relative Reduction Post-Leave Return-to-Work Rate 57.0% US National Average. 90.0% to 94.0% Return Adherence Pregnancy Without Assisted Tech Requires IVF/IUI initiation. 30.0% natural conception rate among fertility members Employer Return on Investment (ROI) Uncoordinated point-solution loss. 2x to 4x validated return ratio This integrated approach contrasts with traditional benefit models like Progyny. While Progyny operates primarily as a fertility benefits manager focused on structured insurance coverage and clinical partnerships, Maven functions as a continuous virtual clinic. By combining digital tools like Maven Wallet for expense reimbursement with clinical resources like Maven Milk, the platform coordinates care across the entire parenting journey. Supported by over $425 Million in venture funding from investors such as General Catalyst, Sequoia, and Dragoneer, Maven has expanded its model into direct-to-consumer (DTC) services. This expansion includes dedicated programs for GLP-1 prescribing, hormone replacement therapies and midlife health, demonstrating how platforms built on employer benefits can expand into consumer-driven markets. Direct-to-Consumer Biomarker Ecosystems: Deciphering the Cash-Pay Longevity and Preventive Space The shift beyond traditional prescriptions has fuelled a growing direct-to-consumer health diagnostic market. Increasingly, health conscious consumers are seeking direct access to advanced medical testing without waiting for symptoms to appear or navigating insurance constraints. Startups like Function Health, Levels, and January Health are building cash-pay models that treat deep diagnostic tracking as a personal wellness investment. Function Health, co-founded by Dr. Mark Hyman in 2023, has emerged as a key player in this direct-to-consumer market. Operating entirely outside the traditional insurance system, Function charges an annual subscription fee (typically $499 billed upfront) to provide members with extensive, regular biomarker testing. To complete this extensive panel, members must follow a strict preparation protocol designed to prevent interference with sensitive biomarker assays: 72 Hours Prior: Stop all vitamins and nutritional supplements, as ingredients like biotin can skew hormone and thyroid assays, though prescribed medications must be continued. 48 Hours Prior: Avoid eating seafood to prevent temporary spikes in mercury measurements. 8 Hours Prior: Fast completely from food and drinks, including black coffee, with the sole exception of water. Morning of Visit: Drink at least 1 litre of water to ensure proper hydration for faster blood draw flow, and avoid physical exercise until the blood draws are complete. Once collected, the raw biomarker data is normalised and uploaded to a HIPAA-compliant digital dashboard. The platform organises results into logical groupings (eg., advanced lipid panels measuring ApoB, hs-CRP, and Lp(a); metabolic markers; thyroid profiles; and hormone levels) and visually maps them against optimal, rather than merely conventional, clinical ranges. Approximately 28 days after testing, members receive a clinical summary from a licensed provider. This report synthesises their biomarker data with their intake goals to deliver personalised diet, lifestyle, and supplement protocols (such as customised list cards detailing top foods to enjoy and avoid). Subscription Component Associated Financial Costs Included Biomarker & Imaging Services Annual Function Membership $499 billed upfront annually. 160+ biomarkers tracked over two semi-annual testing rounds. Out-of-Pocket Lab Fees Approx. $200 per draw visit ($400 total). Paid directly to lab networks for drawing and processing blood. Extended Autoimmunity $249 optional add-on panel. Targeted biomarkers for autoimmune disease detection. Lyme Disease Panel $549 optional add-on panel. Advanced Lyme antibody screening. Celiac Screening $69 optional add-on panel. Immunological markers for gluten sensitivity. Advanced Imaging (MRI/CT) $200 member credit (regularly $999). Whole-body cancer, stroke, and aneurysm imaging. The operational logistics of direct-to-consumer diagnostics require close integration with national laboratory networks. Function Health partners with Quest Diagnostics to process its extensive panel, which requires drawing approximately 35 milliliters of blood. To make this volume manageable for patients, the initial draw is split across two appointments scheduled within a 10-day window. Each visit requires drawing 10 vials of blood, and the second visit includes a urine sample collection. To protect the discounted rates negotiated with its lab partners, Function blocks members from syncing their results with standard laboratory patient portals, hosting all longitudinal trends directly within its own proprietary application. This direct-access testing model is supported by significant venture capital. Function Health has raised over $50 Million from major investors including Andreessen Horowitz (a16z), while its direct competitor Superpower secured a $30 Million Series A round led by Forerunner. These platforms are also integrating emerging diagnostic technologies, such as Verséa Health’s mescreen™, which evaluates cellular energetics and mitochondrial function to track metabolic health at a cellular level. However, this rapid growth has sparked debate within the clinical community. Traditional primary care physicians often argue that testing over 100 biomarkers in asymptomatic individuals inevitably flags minor, clinically insignificant out-of-range values. This "over-testing" can cause patient anxiety, drive unnecessary follow-up evaluations and lead to self-prescribed, unmonitored supplement regimens. Furthermore, critics note that automated clinician notes and generic dietary recommendations can feel formulaic, sometimes lacking the clinical depth of a comprehensive in-person medical assessment. However, proponents maintain that the primary value of these platforms lies in tracking longitudinal biomarker trends over time. By identifying subtle changes within an individual's unique historical baseline, users can make proactive, data-backed lifestyle changes to prevent metabolic and cardiovascular decline before chronic diseases develop. The Regulatory Frontier: Evolving Compliance Requirements and Legal Safeguards As direct-to-consumer wellness devices, health software, and direct-access lab services proliferate, they are navigating a rapidly shifting regulatory landscape. In early 2026, the U.S. Food and Drug Administration (FDA) released updated guidance documents designed to clarify the boundaries between regulated medical devices and low-risk general wellness tools. The 2026 General Wellness Policy Released on January 6th, 2026, the final guidance clarifies how the FDA regulates wearable products that estimate physiological measures. Under this framework, software functions and wearable products intended solely for maintaining, encouraging, or tracking healthy lifestyles, without diagnosing or treating specific conditions are generally exempt from medical device regulation. However, if a device makes specific disease claims (such as diagnosing hypertension or sleep apnea), it crosses the threshold into regulated software or medical hardware, triggering strict review processes. This policy encourages wellness innovation while maintaining oversight for products making diagnostic claims. Clinical Decision Support (CDS) Software Guidance Updated on January 29, 2026, the CDS Final Guidance clarifies which software functions are classified as regulated medical tools. Under this update, software that provides clinicians or consumers with health recommendations must be transparent. If the software operates as a "black box", where the user cannot independently review the underlying data, clinical evidence, or logic behind a recommendation, it is regulated as a medical device. This forces digital platforms (including ŌURA’s diagnostic algorithms and Function Health’s clinical summaries) to clearly explain their data sources and scientific rationale, protecting consumers from unregulated, automated medical advice. AI Predetermined Change Control Plans (PCCP) and Security Protocols The regulatory framework also addresses artificial intelligence integrations and cybersecurity requirements: PCCP Final Guidance (August 2025): Establishes guidelines for Predetermined Change Control Plans (PCCP) in AI-enabled device software. This framework allows developers to implement pre-approved machine learning updates and algorithm modifications without requiring a new 510(k) submission, provided the modifications do not alter the device’s core diagnostic intent. FDA Cyber-Security Requirements (February 2026): Mandates that all connected medical devices and wellness platforms processing patient-generated health data (PGHD) must implement end-to-end data encryption and strict access controls to protect user privacy. Enforcement Actions: The FDA has demonstrated a willingness to enforce these standards, issuing warning letters to wearable manufacturers that make diagnostic disease claims without clinical evidence or regulatory clearance. These evolving rules show that while the FDA is relaxing regulations for low-risk wellness software and consumer devices, it continues to enforce strict oversight on clinical diagnostics and automated decision-making platforms. Wellness startups must design their products within these clear boundaries to ensure consumer safety, maintain clinical credibility and avoid regulatory compliance issues. Nuanced Conclusions and Strategic Outlook The shift from reactive "sick-care" to proactive healthspan management represents a major reorganization of the healthcare market. Traditional healthcare systems are struggling with physician shortages and rising operational costs, leaving consumer technology platforms, virtual clinics and retail pharmacies to build a new model of continuous, daily health support. This structural evolution is driven by consumer demand for diagnostic access and personalised health insights, combined with employer efforts to manage rising healthcare benefit costs. However, the rapid growth of this cash-pay wellness market highlights a deepening division within the healthcare system. While high-income consumers use continuous tracking, advanced biomarkers, and personalised coaching to optimise their health, lower-income and rural populations remain vulnerable to a declining primary care infrastructure. To address these challenges, the healthcare industry must focus on building cohesive, integrated solutions: Standardised API Interoperability: Connecting consumer-generated health data from wearables directly to electronic health records (EHRs) to give primary care providers a continuous, longitudinal view of patient health. Transparent Clinical Support: Developing clear, validated decision support software that translates continuous biometric trends into actionable insights, helping clinicians provide targeted care. Asset-Light Community Access: Repurposing retail spaces to support clinical research, health screenings, and preventive services, ensuring diverse populations can access clinical innovations. Integrated Metabolic Care: Linking physical pharmacy dispensing with food-as-medicine programs, helping patients successfully navigate long-term lifestyle changes. Ultimately, moving "beyond the prescription" requires integrating consumer-driven technology with traditional clinical workflows. By connecting continuous biometric data, accessible diagnostic platforms, and personalised behaviour modification, the healthcare ecosystem can transition from a reactive model of symptom suppression to a proactive, integrated framework focused on extending human healthspan. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

Passive Continuous Phenotyping and Remote Therapeutic Monitoring: The ŌURA Integration Framework Clinical Philosophy of Passive Continuous Phenotyping Passive continuous phenotyping represents a paradigm shift in modern clinical medicine, transitioning diagnostic and therapeutic monitoring from reactive, episodic clinical visits to continuous, proactive and ecologically valid tracking of individual physiological baselines. Traditional medical models rely on static, sparse measurements that are highly vulnerable to clinical artifacts, such as "white-coat" hypertension or retrospective recall bias and fail to capture the dynamic, circadian fluctuations of chronic pathologies. By deploying low-profile consumer biosensors, clinical systems can acquire highly dense, longitudinal physiological data streams directly from a patient’s daily environment without imposing a cognitive or behavioral burden. This continuous acquisition is vital because early physiological decompensation often manifests as sub-clinical deviations in resting metrics, autonomic tone and thermal regulation. Continuous multi-parametric tracking allows for the construction of highly individualised homeostatic baselines. Rather than evaluating an individual against broad, heterogeneous population level norms, clinical algorithms can model intra individual variability to detect micro-deviations that precede acute symptomatic expression. The mathematical modelling of these baseline shifts serves as a predictive early-warning system across multiple domains, including infectious disease onset, cardiovascular degradation and neuropsychiatric status changes. When integrated into clinical care networks, these continuous biometrics enable a proactive care model where therapeutic adjustments are driven by real-world physiological signatures. Biosensing Architecture and Multi-Parametric Clinical Validation The physical ring platform utilises a highly optimised multi-sensor array designed to continuously capture biometrics from the digital arteries on the palmar surface of the finger. This sensing suite comprises three primary components: Dual-Wavelength Infrared Photoplethysmography (PPG): Emits light into the digital arteries to monitor arterial volume changes, tracking pulse waveforms, heart rate, inter-beat intervals (IBI) and pulse transit times. Negative Temperature Coefficient (NTC) Thermistors: Measures skin temperature directly from the finger, capturing continuous peripheral thermoregulatory fluctuations. Tri-Axial Accelerometer: Registers physical movement across three axes to determine sleep-wake cycles, physical activity metrics, and resting periods. To establish clinical utility, these continuous sensor streams have undergone extensive validation against gold-standard clinical measures. This validation spans sleep architecture, autonomic balance, cardiovascular aging, thermal regulation and respiratory disorders. Clinical Outcome Area Reference Standard Ring Performance Metrics Clinical Relevance Source Sleep Staging (Legacy Algorithm) In-lab Polysomnography (PSG) TST underestimated by 32.8\47.3 min; N3 sleep overestimated by 31.5\46.8 min; REM sleep underestimated by 12.8\19.5 min Broadly comparable to research-grade actigraphy (Actiwatch 2). various Sleep Staging (Deep Learning Algorithm) Polysomnography (PSG) 84 overall staging accuracy; REM sleep sensitivity at 90.6 % Light sleep sensitivity at 75.5% High-resolution tracking of longitudinal sleep architecture. various Autonomic Control (HRV/RMSSD) Shimmer3 Electrocardiogram (ECG) Low mean bias in overnight average test; high Pearson correlations for RMSSD and SDNN. Real-world tracking of chronic stress and cardiovascular strain. various Vascular Age (NUS Pipeline) Clinical Fingertip PPG Sensor Mean estimation error of 6\7 years; strong correlation with baseline clinical blood pressure. Scalable, home-based cardiovascular aging assessment. various Hypertension Screening Self-reported status / Cuff measurements 62% sensitivity; 91% specificity; trained on $300,000+$ active study participants. Non-invasive population-level screening for cardiovascular risk. various Nocturnal Dipping Behavior 48-Hour Ambulatory Blood Pressure 84% sensitivity; 69% specificity; Area under the Curve (AUC) of $0.87 Identification of cardiovascular dipping vs. non-dipping risks. various Moderate-to-Severe Sleep Apnea Type 1 In-lab Polysomnography 76% sensitivity; 89% specificity; evaluated on 339 active clinical subjects. Continuous home-based triage of suspected sleep-disordered breathing. various Infectious Fever Monitoring Self-reported Symptom Onset (TemPredict) Changes in daily min temp visible 2\3 days prior to symptom onset in 76\% of COVID-19 cases. Early detection and self-isolation guidance in health systems. various In sleep medicine, early ring models exhibited systematic deviations compared to polysomnography, including an underestimation of total sleep time (TST) and light sleep, alongside an overestimation of deep slow-wave sleep (N3). The deployment of a deep-learning sleep-staging algorithm addressed these gaps, achieving an 84% overall agreement with PSG, with particularly high sensitivity for REM sleep (90.6%). This validation supports its clinical use as an alternative to actigraphy for tracking longitudinal sleep architecture, sleep onset latency (SOL), and wake after sleep onset (WASO). In home-based clinical validation trials against continuous Shimmer3 ECG monitors, the ring demonstrated highly accurate tracking of both nocturnal resting heart rate and RMSSD. Moderate-to-high correlations were also observed for the standard deviation of NN intervals (SDNN) and the percentage of successive normal beat-to-beat intervals differing by more than 50 Milliseconds (pNN50), validating the platform for tracking chronic stress, cardiorespiratory fitness, and autonomic neuropathy. To assess cardiovascular health, researchers from the Centre for Sleep and Cognition at NUS Medicine developed an independent analytical pipeline using overnight peripheral PPG waveforms. By extracting pulse features passively during sleep, their deep-learning model estimated vascular age with a mean error of six to seven years, showing a strong correlation with baseline blood pressure. Continuous skin temperature tracking provides further clinical insight into thermoregulatory homeostasis. Under the TemPredict initiative, co-led by Ashley Mason at the UCSF Weill Institute for Neurosciences and Benjamin Smarr at UCSD, investigators analysed data from over $65,000 international participants. Supported by funding from #startsmall and the Department of Defense (including $800K and $5.1M contract awards), the study demonstrated that the ring's temperature sensor detected early illness patterns in 76% of COVID-19 cases up to 2 to 3 days prior to symptom onset. This predictive capability relied on identifying disruptions in the user's typical circadian temperature minimums and maximums. Additionally, the study revealed a significant correlation between elevated average dermal temperatures, compressed 24-hour temperature fluctuations (reduced circadian amplitude), and the severity of depressive symptoms. This thermal signature indicates that disruptions in peripheral heat dissipation and circadian amplitude may serve as objective biomarkers for major depressive phases, supporting heat-based clinical interventions designed to trigger biological self-cooling mechanisms. Building on these findings, the Oura Science Team released clinical validation data in June 2026 across three key diagnostic domains. First, utilising data from the Blood Pressure Profile Study (which enrolled over 300,000 consenting participants in Oura Labs starting in December 2025), a software-based hypertension screening algorithm demonstrated 62% sensitivity and 91% specificity in identifying self-reported hypertension without a physical cuff. Second, an algorithm designed to classify nocturnal blood pressure dipping behaviours against 48-hour ambulatory references demonstrated 84% sensitivity and 69% specificity, achieving an Area under the Curve (AUC) of 0.87. Third, an algorithm trained to screen for moderate-to-severe sleep apnea against Type 1 attended in-lab polysomnography in 339 subjects demonstrated 76% sensitivity and 89% specificity. This component-based regulatory strategy, following the precedent set by the ring's integration with the FDA-cleared Natural Cycles application for fertility tracking, aims to establish clinical-grade diagnostic pathways on consumer hardware. Remote Therapeutic Monitoring and the Regulatory Reimbursement Landscape Remote Therapeutic Monitoring (RTM) was established by the Centers for Medicare & Medicaid Services (CMS) to reimburse healthcare providers for the systematic tracking of non-physiological data related to treatment adherence, therapy response, and behavioural health outcomes. This model is distinct from Remote Patient Monitoring (RPM), which is restricted to the collection of physiological data (such as blood pressure or blood glucose) using medical devices. RTM explicitly allows for the tracking of subjective, patient-reported outcomes alongside device-derived, non-physiological metrics, including therapeutic exercise adherence, medication compliance, and cognitive behavioural therapy (CBT) engagement. Crucially, while RPM billing is restricted to physicians and non-physician practitioners (such as Nurse Practitioners and Physician Assistants), RTM billing is also open to physical therapists (PTs), occupational therapists (OTs), speech-language pathologists (SLPs), and clinical psychologists. Integrating the ring into an RTM workflow allows providers to monitor therapy adherence and physiologic responses dynamically. For example, in behavioural health and CBT-I (Cognitive Behavioural Therapy for Insomnia) protocols, patients utilise the ring to passively track sleep efficiency, latency, and fragmentation alongside mobile application logs tracking mood and sleep hygiene compliance. The clinical team reviews these consolidated data streams remotely. Cumulative data review time and subsequent tele-support consultations are billed under CPT codes 98980 and 98981, requiring a minimum of 20 minutes of treatment management and at least one synchronous interactive communication per calendar month. This creates a sustainable, reimbursable clinical workflow that supports therapeutic adjustments and patient compliance between standard clinical visits. In addition, RTM billing codes allow for multi-provider clinical coordination. Unlike RPM programs where only a single medical provider can bill for monthly device monitoring, the RTM framework allows multiple specialised clinicians to concurrently track different therapeutic domains. For example, a sleep medicine specialist can monitor a patient's CPAP device and subjective sleep metrics under CPT code 98976, while the patient’s primary care physician manages chronic musculoskeletal pain or metabolic factors using the ring under CPT code 98977. This multi-provider model increases practice revenue and coordinates care across specialties without causing reimbursement conflicts. Enterprise Architecture and Developer Integration Frameworks Building continuous pheno typing pipelines requires a secure, performant and scalable technical architecture to handle high-throughput, time-series data. The integration framework must process raw metrics while ensuring strict data privacy and compliance with global health regulations. The REST API V2 operates at the base URL https://api.ouraring.com/v2. Accessing these data streams requires transition from deprecated Personal Access Tokens (fully deprecated in December 2025) to the standard OAuth 2.0 Authorisation Code flow. The OAuth process redirects users to the consent portal at https://cloud.ouraring.com/oauth/authorize with requested scopes, returning a temporary authorisation code that the backend exchanges at /oauth/token for a 30-day access token and refresh token. Users grant permissions across granular scopes: email, personal, daily (for sleep, readiness, and activity summaries), heartrate (5-minute interval resting and active streams), workout (workout sessions), tag (user annotations), session (guided breathing/meditation), and spo2Daily (average sleep oxygen saturation). If a user's subscription expires, the API returns a 403 Forbidden error, prompting the application to manage account status flows gracefully. For integration testing, developers can call the sandbox endpoint at /v2/sandbox/usercollection/sleep to retrieve mock payloads without connecting live physical rings. Rather than relying on periodic REST polling, which can trigger rate limits (capped aggregate limit of 5,000 requests per 5 minutes), developers can deploy a real-time webhook architecture. When a user opens their mobile application and syncs their ring, raw physiological metrics are pushed to the cloud. Approximately 30 seconds later, the webhook triggers, posting a payload containing the updated resource indicators to the registered callback URL. Webhooks are configured via POST /v2/webhook/subscription. During registration, the endpoint must resolve a GET verification challenge within 10 seconds to activate. The receiving backend should verify the cryptographic HMAC-SHA256 signature transmitted in the x-oura-signature header using its client secret to prevent spoofing. Local testing setups often route these webhooks through tools like ngrok, although developers frequently encounter 504 Gateway Timeout errors during portal registration due to misconfigured challenge-response handlers. To simplify development, teams can utilise pre-built low-code automation tools, data integration toolkits, and unified healthcare APIs. n8n Workflow Nodes: Provides visual integration triggers to retrieve activity, readiness, and sleep summaries, bypassing custom API clients and using HTTP Request nodes for direct REST calls. Tiny Command Recipes: Supports visual workflow recipes to map actions like Get Profile or Get Activity Summary directly into communication channels (such as Slack alerts) or clinical logging targets (Google Sheets, Notion databases). Terra API: Standardises and pushes normalised payloads (Activity, Body, Daily, and Sleep) containing hypnograms and 5-minute heart rate samples to destinations including MongoDB, SQL, Webhooks, Google Cloud Storage, or Firestore. Open Wearables Framework: Standardizes biometrics from multiple providers into a unified schema, managing the token lifecycle and generating auditable metrics like the Resilience Score (which combines HRV and resting heart rate). The Open Wearables framework also supports standard Model Context Protocol (MCP) servers, enabling Large Language Models (such as Claude or ChatGPT) to query clinical data. Rather than analysing raw time-series numbers directly, the MCP server provides pre-computed health scores and anomalies, allowing clinical chatbots and virtual clinics to interpret patient trends against their baseline. Finally, these normalised biometric streams can be routed to electronic health record systems through SMART-on-FHIR pipelines, such as Validic’s Epic App Orchard integration. This middleware enables custom clinical exception rules. Clinicians do not have to review raw sleep data; instead, the system triggers an alert in their Epic InBasket only when physiological metrics (such as sleep fragmentation or resting heart rate) deviate from established safety thresholds over a set period, optimising clinical attention. Passive Continuous Phenotyping and Remote Therapeutic Monitoring: The ŌURA Integration Framework Strategic Deployment Models in Payer and Corporate Systems To drive enrolment and reduce clinical costs, healthcare insurers, employers, and benefits managers are integrating these wearable frameworks into their wellness and chronic care benefits programs. In the payer sector, Medicare Advantage plans (covering over 31 million Americans) leverage wearable metrics to improve Star Ratings and lower acute care claims costs. For instance, Essence Healthcare began a large-scale deployment in 2025, distributing free rings to its Medicare Advantage members. This program aligns patient incentives (free hardware) with provider revenue (RTM billing codes) and payer cost containment (fewer preventable acute hospitalisations). Similarly, private health management programs, such as Discovery Vitality, offer members up to 25% premium discounts when they share verified sleep and physical activity metrics. These continuous biometrics allow insurers to run targeted wellness programs for diabetes, chronic pain, and hypertension, using baseline shifts to prompt clinical outreach. Corporate employers also leverage these platforms to manage group health insurance premiums and reduce absenteeism. Oura for Business provides aggregated wellness and recovery dashboards that correlate sleep deficits with organizational safety risks. Enterprises often offer $200 to $300 hardware credits through integrations with wellness platforms like Virgin Pulse or Wellable to reduce program friction. To navigate regulatory boundaries, these programs must comply with Equal Employment Opportunity Commission (EEOC) guidelines. Updated guidance dictates that all corporate wellness programs must be strictly voluntary; while wellness incentives remain permissible, employers cannot penalise, isolate, or raise premiums for employees who choose not to share their personal biometric data. Systemic Vulnerabilities and Future Engineering Horizons While continuous phenotyping offers substantial clinical utility, developers and systems engineers must address structural vulnerabilities to ensure patient safety and data integrity. The sync latency and "night gap" vulnerability To optimise its small battery and protect user sleep, the ring's Bluetooth transmitter is largely deactivated during sleep. Physiological data does not sync continuously throughout the night; instead, it is cached locally on the ring's onboard memory and transmitted only when the user wakes up and opens the companion mobile application. This creates a systemic "night gap" that limits the device's utility for real-time clinical monitoring. Consequently, the platform cannot support active, real-time emergency alerting (e.g., immediate nocturnal cardiac arrest, sleep apnea, or fall detection). If a life-threatening physiological event occurs, the clinical backend will not receive the data until a manual sync is performed the following morning.\ The clinical ambiguity of missing data A key challenge for clinical tracking is the handling of missing data. Biometric data gaps stem from both physiological and behavioural sources: Motion Artifacts: Optical PPG signals are sensitive to motion noise during sleep, which can disrupt continuous heart rate and HRV recording. Compliance Drops: Users may forget to wear the device or fail to recharge the battery. Zero-Value Ambiguity: Unlike clinical-grade equipment that logs explicit error codes, consumer wearables typically handle missing data by omitting the time-series entry entirely rather than recording a zero value. For automated decision-support systems, distinguishing between a disconnected ring, a dead battery, and actual patient distress is highly complex. If a patient's data stream halts abruptly, clinical algorithms cannot determine if the user has removed the device or experienced a cardiac event. This ambiguity requires developers to implement sophisticated statistical imputation methods to handle sparse records without generating false-positive clinical alerts. Future engineering plans must focus on low-power background sync APIs to reduce sync latency while protecting battery life. Standardising data schemas and error logging will also help clinical backends distinguish technical dropouts from physiological emergencies. As consumer wearables secure further regulatory approvals and integrate with clinical workflows, addressing these structural vulnerabilities will be essential to establishing them as safe, reliable tools for continuous patient monitoring. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk

Strategic Reconfiguration of Global Healthcare: Analysing Prosus’s €400 Million Investment in Alan and the Broader Naspers AI Ecosystem Moat The global technology investment landscape experienced a significant consolidation in June 2026 when the Netherlands-listed technology investor Prosus finalised a €400 Million (US$460 million) direct investment into the French digital health and insurtech platform, Alan. This transaction was executed as the leading component of a larger €480 Million (US$550 Million) Series G funding round. It valued the Paris-based digital insurer at €5.5 Billion (US$6.3 Billion). The investment architecture utilised a structured combination of newly issued primary equity and the acquisition of secondary shares, allowing early liquidity for select historical stakeholders while injecting substantial fresh balance-sheet capital to drive international expansion. This transaction represents an extraordinary valuation step-up within a remarkably compressed operational timeframe. Only three months prior, in March 2026, Alan completed an initial Series G primary capital raise of €100 Million (US$116 Million) at a post-money valuation of €5.0 Billion. The subsequent June financing round, led by Prosus, elevated Alan's market capitalisation by €500 Million in less than 90 days. This trajectory highlights strong investor demand and highlights a fundamental re-rating of Alan’s underlying operating model. The broader investment syndicate for this round involved both elite venture capital and sovereign-scale allocations. Prosus was joined by existing institutional backers Teachers' Venture Growth (TVG) and Index Ventures, alongside a new equity commitment from Dara Holdings. Operationally, the final closing of the round remains subject to standard regulatory clearances, primarily from the French Prudential Supervision and Resolution Authority (ACPR), which serves as Alan’s direct insurance supervisor. Financial and Transactional Milestones Transaction Details Strategic and Market Implications Series G (Primary Phase) March 2026; €100 million raised at a €5.0 billion post-money valuation Established the baseline valuation and provided initial capital for international market entry. Series G (Consolidated Phase) June 2026; €480 million total round size led by a €400 million commitment from Prosus Valued the company at €5.5 billion ($6.3 billion); introduced secondary liquidity and primary capital. Cumulative Funding to Date Approximately $1.3 billion across nine structured funding rounds Positions Alan among the most capitalized digital health platforms globally. Q1 2026 ARR Performance €800 million (US$909 million), representing 53% year-on-year growth Demonstrates the rapid transition of the business from a local startup to an institutional scale. This funding round occurred amid a broader shift in the European digital health and insurance technology sectors. The total European private health insurance market was valued at approximately $517 Billion in 2025 and is projected to expand to $772 Billion by 2034. Private capital has responded to this market size: European digital health companies raised $1.2 Billion in venture funding in the first quarter of 2026 alone, with digital health insurance platforms capturing $116 Million of that total. This trend is further illustrated by the rapid close of dedicated institutional vehicles, such as Lauxera Capital’s €520 Million healthtech fund, which closed in under 18 months, indicating strong investor confidence in late-stage European healthcare platforms. Alan’s capital strategy is designed to establish "prevention insurance" as a global category. Part of the Series G capital will fund expansion beyond France into Spain, Belgium, and Canada. To build brand equity as it enters these highly regulated markets, Alan secured a partnership with international football star Kylian Mbappé, who invested in the company and signed on as a global brand ambassador. Additionally, the company is targeting new demographics to sustain its growth. It plans to launch "Alan Campus" in January 2027, a digital-first health insurance product designed for students aged 18 to 28. This offering aims to capture users early in their healthcare journeys and transition them into long-term enterprise contracts as they enter the workforce. The Operational and AI Architecture of Alan: Unlocking Extreme Operating Leverage The strategic logic of Prosus’s investment lies in Alan’s capacity to decouple policyholder growth from administrative headcount. Traditional health insurance operations scale linearly: as policyholder numbers grow, insurers must expand their back-office teams to handle claims, underwriting, and customer support. In contrast, Alan serves more than 1.1 Million members and 37,000 corporate clients with a lean workforce of only 850 employees. This structural efficiency is enabled by its AI-native platform, which automates core operational workflows. This automated operational architecture has a significant impact on unit economics. By deploying large language models to automate claims processing, fraud detection, and customer support triage, Alan reduced its per-member administrative costs by 28% in 2023. Because its infrastructure is digital-first, most standard reimbursements are processed, verified, and disbursed to policyholders within 24 hours. A key factor in Alan's technical positioning is its relationship with Paris-based Mistral AI. Both of Alan's co-founders, Jean-Charles Samuelian-Werve (CEO) and Charles Gorintin (CTO), serve as co-founding advisors and board members of Mistral AI, which was established in 2023. This relationship provides Alan with direct access to advanced open-source large language models. It allows the company to integrate specialized, highly secure clinical and administrative language models directly into its insurance platform while maintaining strict compliance with European data sovereignty regulations. This integration of AI into a regulated business model creates an interesting valuation dynamic. While Alan uses AI to power much of its platform, the vast majority of its revenue still comes from traditional insurance premiums on collective corporate contracts. This shows that while Alan is structurally a regulated insurance business, its AI-driven efficiencies allow it to achieve the high operating leverage and margin profiles typically associated with pure-play software companies. This operational profile is what justifies its premium, AI-like valuation multiples in private markets. The NasperscProsus Portfolio Strategy: Transitioning to an Integrated Platform Play The investment in Alan is part of a broader strategic shift by Prosus and its parent company, Naspers. In 2019, Naspers spun off and listed its international internet assets as Prosus on the Euronext Amsterdam. This move was designed to unlock shareholder value and address the persistent valuation discount of Naspers shares relative to its underlying 2001 investment in Chinese technology giant Tencent. Historically, Naspers focused on early-stage, region-specific venture investing through vehicles like Naspers Foundry. This initiative deployed R1.4 Billion (approximately $300 Million) into South African startups, backing businesses such as home services platform SweepSouth, agritech platform Aerobotics, and early insurtech players Ctrl and Naked Insurance. Under current CEO Fabricio Bloisi, however, the group has shifted toward a more integrated global platform strategy. Prosus is actively working to transition its portfolio of consumer internet businesses into an interconnected "life assistant" platform. This strategy aims to integrate food delivery, payments, fintech, travel, and classifieds into a single consumer ecosystem. To support this unified ecosystem, Prosus entered a global partnership agreement with Amazon Web Services (AWS) in early 2026. This agreement standardises artificial intelligence infrastructure across Prosus's global operations, allowing portfolio businesses, including iFood, OLX, PayU, Despegar, eMag, and Just Eat Takeaway, to build on shared technical frameworks. By investing in Alan, Prosus adds a highly scalable, digital health and preventative care layer to this ecosystem. This healthcare integration allows Prosus to connect its transactional platforms, such as food delivery and wellness commerce, with a regulated medical and insurance network. Investment Platform or Vehicle Core Mandate & Geographic Focus Historical / Current Operational Scaling AI Integration & Ecosystem Role Naspers Foundry Early-stage, South Africa-focused venture capital Deployed over R200 million across key initial holdings like SweepSouth and Aerobotics. Addressed local socioeconomic needs via localized technology applications. Prosus Ventures Global early-stage and growth technology investments Active globally; committed over $400 million across 40+ deals in FY25. Backs early-stage AI startups that optimize logistics, commerce, and healthcare workflows. AWS Global Partnership [ Enterprise cloud and AI standardization framework Standardised cloud operations across major international business units Leverages AWS advanced machine learning to build secure, compliant consumer experiences. Technical Analysis of the Prosus AI Engine: Large Commerce Models and Agentic Infrastructure At the core of Prosus’s technology strategy is its proprietary Large Commerce Model (LCM). The LCM is a specialized, agentic AI model trained on Prosus's global database of commercial transactions. This database comprises over 10 Trillion tokens of commercial data, reflecting real-world interactions from 500 Million users and 5 Million merchant partners. Unlike general-purpose large language models, the LCM is built to understand, reason about and anticipate consumer transactional intent. It can interpret complex, multi-variable requests and execute actions across different transactional platforms. The operational efficiency of the LCM is a major differentiator. In benchmarking tests, the model proved to be 60 times cheaper to run than leading general-purpose frontier models while delivering higher conversion accuracy on commercial and transactional tasks. Prosus has demonstrated the model's commercial impact through its food delivery business, iFood, in Brazil. The platform has deployed the LCM across more than 40 distinct use cases, resulting in a 19% increase in home-screen conversions, a 51% reduction in advertising acquisition costs, and a 75% increase in push-notification engagement. To bring this agentic capability to its broader partner network, Prosus launched ToqanClaw and Zapia. ToqanClaw is a secure, no-code development platform integrated with Prosus's internal AI engine, Toqan. The system allows businesses—such as restaurants, merchants, and logistics partners—to build custom automations and analytic dashboards using natural language prompts. To optimize performance and minimise costs, ToqanClaw dynamically routes tasks across more than 20 open-source and proprietary language models. This strategy has reduced operational token costs for participating small businesses by up to 90%. Platform / Tool Core Technical Architecture Performance & Cost Benchmarks Consumer / Enterprise Deployment Large Commerce Model (LCM) Proprietary agentic AI model trained on 10 trillion commercial tokens 60 times cheaper to run than leading general-purpose frontier models Deployed across iFood, OLX, Just Eat Takeaway, and eMag. Toqan / ToqanClaw Secure, multi-model no-code development platform Reduces token execution costs by up to 90% via dynamic routing Deployed to over 5 million merchant partners and small businesses. Zapia Consumer-facing, autonomous agentic AI assistant Executes multi-step, real-world tasks across messaging APIs Reached 7 million monthly active users, primarily in Latin America. Zapia is the consumer-facing interface for this agentic ecosystem. Rather than acting as a standard conversational chatbot, Zapia is an autonomous assistant designed to execute multi-step tasks across different apps. For example, a user can instruct Zapia to find a restaurant with specific dietary options, coordinate a booking with friends over WhatsApp, gather their preferences, and confirm the reservation. Currently serving over 7 Million users, Zapia’s growth has been driven by its integration with WhatsApp in Latin America. By connecting Alan’s clinical navigation tools with the LCM, ToqanClaw, and Zapia, Prosus can provide Alan's members with highly automated, conversational healthcare navigation. This allows users to manage wellness tracking, schedule medical appointments, and coordinate insurance approvals within a single, unified interface. Strategic Reconfiguration of Global Healthcare: Analysing Prosus’s €400 Million Investment in Alan and the Broader Naspers AI Ecosystem Moat Mapping Prosus’s Global Healthtech Assets: From Infrastructure Layers to AI Diagnostics The investment in Alan is part of a broader, multi-region healthcare technology portfolio managed by Prosus. The group’s healthcare strategy focuses on platforms that integrate clinical workflows, automate documentation, and streamline healthcare distribution. Mevo (Brazil) In March 2026, Prosus led an $18 Million (R$95 Million) funding round for Mevo, Brazil's leading e-prescription platform. Historically, Brazil's healthcare system processed over one billion handwritten prescriptions annually, leading to high administrative friction and risk of errors. Mevo addresses this by connecting doctors, clinical institutions, patients, and retail pharmacies through a unified digital prescription network. The company operates a transactional monetization model. Its digital prescription system is provided to clinical institutions and physicians at no upfront cost to capture distribution. Monetisation occurs downstream through Mevo Shop, an integrated consumer portal that facilitates exam scheduling, patient support program integration, and direct medicine fulfilment through a network of 11 major pharmacy chains across 1,600 retail locations. Mevo has integrated its platform into more than 1,100 healthcare institutions and seven major health insurance providers, covering approximately 30% of Brazil's private healthcare market. The company projects that it will process digital prescriptions for over 20 Million patients in 2026. Corti (Denmark) Co-led by Prosus Ventures and Atomico, Copenhagen-based Corti secured a $60 Million Series B in September 2023 to scale its clinical AI co-pilot. Corti’s software processes speech and text in real time during clinical consultations and public safety emergency calls. The AI provides diagnostic suggestions, flags critical symptoms (such as cardiac arrest during emergency calls), and automates post-encounter administrative workloads, including clinical coding, documentation, and quality assurance. Corti currently supports clinical workflows over 150,000 times a day, covering approximately 100 million patients annually across Europe and the United States. VOA Health (Brazil) In March 2025, Prosus backed Belo Horizonte-based VOA Health with a $3 Million Seed investment. VOA Health develops generative AI tools designed to streamline clinical documentation for healthcare providers. By automating the creation of medical records, patient histories, and prescriptions, the platform aims to reduce administrative workloads and allow doctors to focus more on direct patient care. PharmEasy (India) Prosus’s investment history in India’s PharmEasy (API Holdings) highlights the financial complexities of scaling digital health platforms in emerging markets. In April 2021, Prosus Ventures and TPG co-led a $350 Million Series E round that established PharmEasy as a unicorn at a valuation of approximately $1.5 Billion, which subsequently climbed to a peak private valuation of $5.6 Billion in late 2021 following the acquisition of listed diagnostics provider Thyrocare. However, post-pandemic macroeconomic tightening and the postponement of its planned public offering forced PharmEasy to undergo severe corporate restructuring. By 2023, the company faced substantial valuation markdowns and debt-servicing challenges. To stabilise its balance sheet, PharmEasy completed a highly dilutive rights issue in late 2023 / early 2024, raising approximately $417 Million (including direct participation from Prosus and Temasek), followed by a $216 Million capital round in April 2024 led by Ranjan Pai (MEMG Family Office) at a restructured valuation of approximately $710 Million. This restructuring was concluded in September 2025, with PharmEasy securing $193 Million (INR 1,700 crore) in debt financing led by 360 ONE Asset to refinance high-cost debt and focus strictly on positive EBITDA generation. Operationally, the company successfully pivoted from a capital-intensive inventory model to a highly profitable, asset-light aggregator network. This transition was achieved by utilizing machine learning models to reduce partner pharmacy stock-outs by 40%, expanding high-margin diagnostic services via Thyrocare's 3,000+ collection centers, and scaling higher-margin private-label wellness products. Portfolio Company Regional Focus & Scope Key Financial Event AI & Automation Drivers Strategic Integration / Status Mevo Brazil; covers 30% of the private healthcare market. Led by Prosus in March 2026; $18 million Series B extension. Digital e-prescription engine and integrated marketplace. Connected to 1,100+ medical centers and 1,600+ pharmacies. Corti Europe & US; processes 150,000 daily patient consultations. Co-led by Prosus Ventures and Atomico; $60 million Series B. Real-time speech and text analysis for emergency triage. Optimizes clinical workflows and automates billing codes. VOA Health [ Brazil; operates in clinical documentation space. Funded by Prosus in March 2025; $3 million Seed round. Generative AI to automate patient histories and clinical records. Focuses on reducing administrative overhead for clinicians. PharmEasy India; serves over 20 million registered users across 1,200+ cities. Series E co-led by Prosus Ventures; secured $193M debt in Sept 2025. ML models for inventory demand forecasting to cut stock-outs. Transitioned to an asset-light aggregator with positive EBITDA. Clarity Global; operates in highly regulated customer support. Early-stage venture investment backed by Prosus Ventures. Secure AI customer experience platform for healthcare/finance. Enhances enterprise data security and compliance. Dognosis Global; focuses on early-stage non-invasive diagnostics. Early-stage funding round in March 2026. Canine olfaction, robotics, and AI to identify cancers from breath. Expands preventative diagnostic capabilities in oncology. Strategic Implications for the Global Managed Care and Insurtech Landscapes The convergence of global capital and specialised AI models across Prosus’s healthcare portfolio suggest several key structural shifts in the healthcare and insurance markets. The Decoupling of Margin Profiles in Managed Care Traditional insurance carriers have historically operated with high administrative loss ratios and low operating margins due to the intensive manual labor required to manage claims processing, clinical reviews, and member support. By using specialized open-source LLMs like Mistral for claim auditing and customer interactions, digital platforms can scale their membership base exponentially while keeping headcount flat. This shifts the financial profile of an insurtech platform from a traditional financial-services multiple to a high-margin software multiple, driving the valuation premium seen in Alan’s €5.5 Billion valuation. Regulatory Protectionism as a Structural Barrier to Entry Unlike standard consumer software, healthcare operates within a highly regulated environment. Insurtech platforms must maintain significant capital reserves and secure approvals from national supervisors like the ACPR in France. This regulatory complexity acts as a powerful barrier to entry for big-tech competitors. Consequently, global technology investors like Prosus must deploy capital into vertically integrated, pre-regulated players (such as Alan) rather than attempting to build alternative consumer-facing digital health layers from scratch. The Evolution from Transactional E-Commerce to Ambient Life Assistants The integration of specialised AI tools like Prosus’s Large Commerce Model with clinical assets like Mevo and Alan indicates a shift in how consumers access healthcare. Instead of navigating multiple disjointed apps for health insurance, doctor consultations, prescriptions, and lifestyle-related purchases, consumer interaction is consolidating around singular, agentic conversational hubs. This agentic orchestration can autonomously manage the entire care loop: tracking a user's preventative wellness goals, flagging clinical risks, booking consultations, authorising insurance claims, and delivering prescriptions directly to the home. This continuous loop optimizes health outcomes while creating a highly defensible, data-rich ecosystem moat around the underwriting platform. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada

The Scaleup Europe Fund: Navigating Late-Stage Growth Capital Opportunities for European HealthTech and MedTech Companies Strategic Genesis: Combatting the European Tech Exodus in Life Sciences Despite Europe's thriving startup ecosystem, which consistently produces world-class scientific discoveries and breakthrough technologies, innovative enterprises face severe challenges when attempting to scale within the continent. As tech and life science companies transition from early-stage validation to the capital-intensive scaleup phase, they routinely encounter an underdeveloped late-stage venture capital sector. This domestic funding gap, particularly acute in growth rounds approaching or exceeding nine figures, has historically forced many of Europe's most promising innovators to seek capital from foreign institutional investors, often relocating their headquarters, executive talent, and intellectual property to the United States or Asia. This capital flight represents a systemic drain on European economic value and technological sovereignty. The vulnerability of European innovation is underscored by high-profile foreign acquisitions of pioneering technology firms. Notable examples include Advanced Micro Devices acquiring the Finnish artificial intelligence company Silo AI for $665 Million in 2024, and IonQ's purchase of the UK-based quantum computing firm Oxford Ionics for more than $1 Billion in 2025. To curb this trend and foster domestic alternatives, European policymakers have designed the Scaleup Europe Fund as a concrete deliverable of the European Commission's Competitiveness Compass and a direct response to the Draghi Report's urgent call to close Europe's deep-tech financing gap. Operating as a key pillar of the EU Startup and Scaleup Strategy, the Scaleup Europe Fund is designed to serve as a strategic autonomy tool. Implemented with the support of the European Investment Bank, the fund acts as a catalyst to mobilise private and public institutional capital, ensuring that the critical technologies of the next decade remain anchored in Europe. For healthtech, medtech and biotechnology companies, sectors characterised by highly capital-intensive development timelines, prolonged clinical trials, and stringent regulatory processes, this fund represents a transformative shift in the European funding landscape. Core Architecture: Fund Size, Structure and Private Governance The Scaleup Europe Fund is positioned to become Europe's largest technology growth fund, targeting €5 Billion in total commitments. To achieve global parity and effectively compete with the massive venture funds of the US and China, the European Commission has signaled intentions to scale the fund's total capacity to €25 Billion over time. The core capital structure is a public-private hybrid. The European Commission has committed €1 Billion to the fund, sourced directly from Horizon Europe, the EU's flagship research and innovation programme. The remaining €4 Billion is being raised from private and sovereign institutional investors, with private partners having already committed €1.5 Billion by the end of 2025. Following a highly competitive public call for expressions of interest launched between December 2025 and February 2026, the European Innovation Council Fund Board shortlisted elite management firms from Sweden, the UK, and France to oversee the vehicle. This shortlist included EQT and Northzone from Sweden, Eurazeo from France, and Atomico and Vitruvian Partners from the United Kingdom. In May 2026, Swedish investment firm EQT was appointed as the preferred adviser and fund manager. EQT operates as an independent, market-based manager, ensuring that investment decisions are commercially driven and free from political influence. Furthermore, EQT commits a significant portion of its own capital to the fund, establishing a profit-making framework that aligns public policy objectives with private sector returns. Structurally, the Scaleup Europe Fund will operate as a separate compartment within the existing EIC Fund umbrella. This architecture allows the fund to deploy significantly larger investment tickets than previous European funding mechanisms. While the EIC Fund historically capped its equity investments at €30 Million, the Scaleup Europe Fund will focus on direct equity investments in the range of €100 Million and above. This capability is designed to lead or co-lead major financing rounds, helping European innovators scale into "unicorns" without relinquishing operational control to foreign buyers. Strategic Alignment: Positioning Healthtech, Medtech and Biotech for Scale For healthtech, medtech, and biotechnology scaleups, the launch of the fund, scheduled to execute its first investments in autumn 2026, addresses the financial mismatch inherent in scaling life science technologies. Healthcare innovation requires patient, long-term growth capital to finance multi-centre Phase II/III clinical trials, secure global regulatory approvals (such as CE-MDR and FDA clearances) and construct advanced, compliant manufacturing facilities. The availability of €100 Million+ investment tickets provides the financial runway required to navigate these hurdles. The composition of the fund’s founding investors provides portfolio companies with strategic benefits that extend far beyond capital injection. Founding Investor Group Institutional Members Strategic Value to Healthtech & Medtech Scaleups Life Sciences Specialists Novo Holdings Brings unmatched clinical, commercial, and industrial expertise, with a deep understanding of biotechnology development and global drug distribution networks. Sovereign & Promotional Banks EIFO (Denmark), Wallenberg Investments (Sweden), BGK (Poland) Provides stable, long-term backing linked to national and European industrial strategies, facilitating local regulatory navigation and commercial procurement within regional health systems. Pension & Asset Managers APG Asset Management (representing Dutch pension fund ABP), Allianz (German Insurer) Represents vast pools of non-dilutive, patient capital capable of supporting successive growth rounds and stabilizing long-term corporate valuation. Commercial Banking Networks Santander/Mouro Capital, Intesa Sanpaolo Offers access to cross-border financial services, debt facilities, corporate banking, and established corporate healthcare networks across Europe. Philanthropic Foundations Fondazione Compagnia di San Paolo, Fondazione Cariplo Bridges the gap to clinical research institutions, university hospital networks, and public health initiatives aligned with broader societal and UN development goals. The management of the fund will actively leverage EQT's specialised internal divisions, particularly its dedicated Life Sciences, Ventures, and Growth teams. This cross-disciplinary expertise is essential for modern healthtech companies, which frequently operate at the intersection of biology, software, and advanced engineering. By connecting scaleups with EQT's global network of clinical advisors, hospital executives, and industrial partners, the fund helps portfolio companies accelerate commercialisation, optimise supply chains and secure market entry across multiple European jurisdictions. Comparative Framework: Navigating the Multi-Tiered European Growth Capital Ecosystem To navigate the expanding European funding landscape, healthcare executives must understand the distinct operational roles of the EIC Accelerator, the EIC STEP Scale Up call, and the Scaleup Europe Fund. Each mechanism represents a different level of maturity, funding capacity, and qualification requirements. Feature / Metric EIC Accelerator EIC STEP Scale Up Call Scaleup Europe Fund (SEF) Target Technology Maturity TRL 6 to 8 (Technology demonstrated in relevant environments) Late-stage development & rapid scaling of strategic tech Series B & C commercial scaling, global deployment, and market expansion Primary Financial Instruments Blended finance: up to €2.5M in grants and €0.5M to €10M (or up to €15M) in equity Equity-only investments Direct equity and quasi-equity growth capital Funding Amount (Ticket Size) Max €2.5M grant; equity capped typically at €10M–€15M €10 million to €30 million €100 million and above Institutional Scale Under the main EIC Fund umbrella Part of the EIC STEP call within the 2026 Work Programme Independent, privately managed compartment under EQT Evaluation Framework Standard EIC multi-stage process with quarterly cutoff dates Quarterly batching system for rigorous strategic and financial proposals Direct application to the fund manager (EQT) utilizing commercial private market due diligence Co-investment Requirements Evaluates risk; crowding in private capital is desired but not rigidly benchmarked at entry Strictly enforces the "20% Rule" pre-commitment from a qualified lead investor Market-driven, leading or co-leading major, syndicated European investment rounds To ensure application readiness, corporate planning must align with the specific batching and evaluation timelines established under the EIC 2026 Work Programme. The cut-off dates for both early-stage Accelerator proposals and the mid-tier STEP Scale-Up program are highly structured, requiring significant lead times for preparation. Evaluation Cycle (2026) EIC Accelerator Step 2 Cut-off Dates EIC STEP Scale-Up Cut-off Dates First Batch / Q1 January 7, 2026 February 11, 2026 Second Batch / Q2 March 4, 2026 May 6, 2026 Third Batch / Q3 May 6, 2026 September 9, 2026 Fourth Batch / Q4 July 8, 2026 November 25, 2026 Fifth Batch / Q5 September 2, 2026 No corresponding cut-off in Q5 Sixth Batch / Q6 November 4, 2026 No corresponding cut-off in Q6 The Scaleup Europe Fund: Navigating Late-Stage Growth Capital Opportunities for European HealthTech and MedTech Companies Geographic Compliance, UK Exclusion and Swiss Bilateral Frameworks As a concrete delivery of European strategic autonomy, the Scaleup Europe Fund mandates that funded companies retain the majority of their value creation, including intellectual property, manufacturing capabilities and headquarter functions, within the European Union or Horizon Europe associated countries. This geopolitical positioning has significant strategic and operational implications for healthtech and medtech companies located in non-EU-27 jurisdictions, specifically within the United Kingdom and Switzerland. The United Kingdom: Complete Equity Exclusion and Corporate Redomiciling The post-Brexit relationship between the United Kingdom and the European Union creates a restrictive operational environment for British healthtech and medtech enterprises. While the UK’s formal association with Horizon Europe allows British SMEs and research consortiums to participate in research centric grants, such as the EIC Pathfinder and EIC Transition initiatives, it explicitly excludes them from accessing any equity or investment components. Consequently, UK-domiciled startups are eligible to apply for the grant portion of the EIC Accelerator (up to €2.5 Million for TRL 6-8 development), but they are completely ineligible for the EIC STEP Scale-Up call and the Scaleup Europe Fund. For high-growth British healthtech scaleups seeking to access these multi-million-euro equity pools, the only viable pathway is to execute a corporate restructure, effectively establishing a primary operating headquarters and transferring key intellectual property to an eligible EU Member State or Associated Country prior to submitting a full application. Switzerland: Bilateral Pathways and Strict Compliance Audits In contrast to the UK, Switzerland's active participation as an Associated Country to Horizon Europe allows Swiss-domiciled startups and "small mid-caps" (up to 499 employees) to qualify for late-stage equity initiatives, including the Scaleup Europe Fund. However, Swiss healthtech scaleups must navigate a rigorous, multi-layered compliance audit to demonstrate their strategic alignment with the broader European ecosystem. Swiss companies cannot pitch their solutions solely within the context of their domestic market. Instead, they must explicitly show how their technologies, manufacturing pipelines, and clinical data structures strengthen European industrial sovereignty and security of supply, particularly for products on the Union list of critical medicines. Furthermore, Swiss applicants must present a funded, executable plan detailing how they will establish or scale physical operations, clinical testing sites, or commercial distribution centres within EU Member States within 12 months of receiving the investment. Executive Playbook: Qualification Benchmarks and Pre-Commitment Rules For healthtech and medtech executives preparing to apply for the Scaleup Europe Fund or the mid-tier EIC STEP program, operational planning must begin six to twelve months in advance. Because the fund is overseen by EQT alongside highly analytical institutional investors, the diligence process focuses heavily on financial discipline, commercial scalability, and regulatory compliance. Operational and Financial Benchmarks To qualify for these late-stage growth vehicles, a healthtech company must demonstrate robust commercial traction. While early-stage grants focus primarily on scientific novelty, the Scaleup Europe Fund and its precursors target scaleups with established product-market fit. The fund typically targets enterprises with annual revenues ranging between €2 Million and €15 Million, with a proven, audited year-over-year growth rate exceeding 25% for at least two consecutive fiscal years. For the EIC STEP Scale-Up program, the commercial benchmarks are even higher, requiring an Annual Recurring Revenue (ARR) of over €10 Million and gross margins exceeding 40%. Although current profitability is not a strict requirement, the application must include an investor-grade, five-year financial model outlining a credible path to EBITDA breakeven within three to four years post-investment. The Pre-Commitment Mandate and the "20% Rule" A critical hurdle for healthtech scaleups is the "20% Rule" pre-commitment requirement mandated under the EIC STEP Scale-Up framework. Companies cannot apply for these public-private growth equity tickets in a vacuum. To secure a €10 Million to €30 Million ticket, the applicant must have already obtained a formal, legally backed commitment from a qualified private lead investor. This lead investor must commit to funding at least 20% of the total target round. Furthermore, the total funding round must be three to five times the size of the requested EU equity ticket. For example, if a medical device scaleup requests a €20 Million equity ticket from the EIC, it must target a total funding round of €60 Million to €100 Million, with a private lead investor committing at least €12 Million to €20 Million. This mechanism ensures that public capital acts as a cornerstone to catalyze larger private co-investments, de-risking the round and accelerating market deployment. Data Sovereignty and Regulatory Auditing Because the fund serves as a strategic autonomy tool, healthtech scaleups must ensure that their technical and regulatory frameworks align with European standards. This is particularly critical for digital health and artificial intelligence companies, where compliance with the EU AI Act and GDPR must be integrated directly into the product design. Investee companies are expected to host their infrastructure on European servers, minimise dependencies on non-EU cloud hyperscalers to mitigate vendor lock-in, and maintain absolute, unencumbered ownership of their core software codebase, patented clinical algorithms, and proprietary medical data. Conclusions and Strategic Recommendations for Board-Level Decision-Makers The establishment of the Scaleup Europe Fund represents a significant evolution in the European Union’s approach to venture capital and industrial planning. By partnering with an experienced global manager like EQT and securing anchor commitments from major institutional investors like Novo Holdings and Allianz, the European Commission is shifting from grant-based, fragmented funding toward a market-driven growth equity model. This structural change provides healthtech, medtech, and biotech scaleups with the large-scale, long-term capital required to build global healthcare leaders without having to leave the European ecosystem. For healthtech and medtech boards of directors, the emergence of this multi-billion-euro funding architecture requires immediate, proactive adjustments to corporate finance and operational strategies: Sovereignty-by-Design in Corporate Structuring: Healthtech scaleups, particularly those based in the UK, must evaluate their long-term corporate geography. If late-stage capital requirements exceed €50 million, boards should consider establishing an operational headquarters, transferring intellectual property, or redomiciling to an eligible EU Member State or Associated Country. Similarly, Swiss entities must structure their growth plans to include rapid, measurable expansion of operations, manufacturing, or clinical sites inside the EU-27 borders. Alignment with Life Science Anchor Backers: Healthcare scaleups should tailor their business models to align with the strategic goals of the fund's founding partners. Positioning technologies to align with the investment criteria of life science specialists like Novo Holdings, or ensuring compatibility with the long-term pension goals of APG/ABP, can significantly improve a company's prospects during the due diligence process. Accelerating Investor Readiness: Scaleup executives should initiate comprehensive pre-audit processes to ensure their financial reporting, beneficial ownership records, and data architectures meet institutional standards. Preparing five-year financial models that demonstrate a clear, audited path to EBITDA breakeven within three to four years is essential to withstand the rigorous due diligence of independent managers like EQT. Securing Qualified Lead Investors: Because the EIC STEP Scale-Up program and the broader fund rely on co-investment structures, companies must build relationships with qualified private lead investors well ahead of application deadlines. Securing a firm commitment for at least 20% of the target round from a private lead investor is a necessary first step to unlock public matching equity. By implementing these strategic preparations, healthtech and medtech enterprises can position themselves to successfully leverage the Scaleup Europe Fund, turning Europe's boldest venture capital initiative into a launchpad for global healthcare leadership. Nelson Advisors > European MedTech and HealthTech Investment Banking Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk Nelson Advisors regularly publish Thought Leadership articles covering market insights, trends, analysis & predictions @ https://www.healthcare.digital Nelson Advisors publish Europe’s leading HealthTech and MedTech M&A Newsletter every week, subscribe today! https://lnkd.in/e5hTp_xb Nelson Advisors pride ourselves on our DNA as ‘Founders advising Founders.’ We partner with entrepreneurs, boards and investors to maximise shareholder value and investment returns. www.nelsonadvisors.co.uk #NelsonAdvisors #HealthTech #DigitalHealth #HealthIT #Cybersecurity #HealthcareAI #ConsumerHealthTech #Mergers #Acquisitions #Partnerships #Growth #Strategy #NHS #UK #Europe #USA #VentureCapital #PrivateEquity #Founders #SeriesA #SeriesB #Founders #SellSide #TechAssets #Fundraising #BuildBuyPartner #GoToMarket #PharmaTech #BioTech #Genomics #MedTech Nelson Advisors LLP Hale House, 76-78 Portland Place, Marylebone, London, W1B 1NT lloyd@nelsonadvisors.co.uk paul@nelsonadvisors.co.uk Meet Nelson Advisors @ 2026 Events Digital Health Rewired > March 2026 > Birmingham, UK NHS ConfedExpo > June 2026 > Manchester, UK HLTH Europe > June 2026, Amsterdam, Netherlands HIMSS AI in Healthcare > July 2026, New York, USA Bits & Pretzels > September 2026, Munich, Germany World Health Summit 2026 > October 2026, Berlin, Germany HealthInvestor Healthcare Summit > October 2026, London, UK HLTH USA 2026 > October 2026, USA Barclays Health Elevate > October 2026, London, UK Web Summit 2026 > November 2026, Lisbon, Portugal MEDICA 2026 > November 2026, Düsseldorf, Germany Venture Capital World Summit > December 2026 Toronto, Canada Nelson Advisors specialise in Mergers and Acquisitions, Partnerships and Investments for Digital Health, HealthTech, Health IT, Consumer HealthTech, Healthcare Cybersecurity, Healthcare AI companies. www.nelsonadvisors.co.uk