Digital health technologies are poised to profoundly transform the healthcare sector. These technologies leverage the data now available from an increasing range of sources including wearables, connected and smart medical devices, increasingly enabling the development of Artificial Intelligence/ Machine Learning (AI/ML) powered diagnostic and interventional tools.
Regulatory bodies, such as the FDA, the MHRA and the EU, are making strides in defining and regulating AI/ML-enabled medical devices, with a focus on ensuring safety and efficacy. Despite this progress, navigating the evolving regulatory landscape poses challenges, especially as digital health solutions can blur the lines between medical devices and non-medical tools.
Application of AI in digital health
More and more AI/ML tools are being developed in the healthcare sector to help manage vast amounts of data and interpret it speedily and accurately. While concerns around potential for harm and bias are beginning to outline possible limitations to the adoption of AI/ML, AI and ML offer immense potential in revolutionizing healthcare delivery, enabling early disease detection, personalized treatment approaches, and remote patient monitoring to name a few.
Whether this data is in text form (such as notes), video or imagery, AI/ML can help save hours of manual analysis and cross-checking and suggest interpretation that would otherwise take human researchers years to complete. AI/ML can rapidly analyse radiology images, histological data, posture, eye movement, speech speed, pitch and sound and a whole range of other types of input.
Some example areas for application of AI/ML within medical devices are Diagnostic Imaging, Remote Patient Monitoring, Personalized Medicine, Clinical Decision Support Systems (CDSS), Wearable Health Devices, Robotic Surgery, Predictive Analytics for Healthcare Management, demonstrating the versatility and potential impact of AI and ML in medical device development and healthcare delivery, spanning from diagnosis and treatment to patient monitoring and management.
In underfunded areas of medical research, this could even prove life changing by helping detect comorbidities, environmental or genetic factors that place particular individuals at higher risk of disease. Moving the bar even higher than early detection, it could become possible to warn people of an estimated potential risk years before diseases begin to manifest.
Regulatory considerations for AI/ML in healthcare technology
From Digital Health to Medical Devices
Distinguishing between digital health products and digital medical devices can be complex but is essential for understanding regulatory nuances. Digital health spans a wide range of technologies, from non-medical devices aimed at monitoring well-being to medical devices designed for specific medical purposes.
The International Medical Device Regulators Forum (IMDRF) provides a generally accepted, though not exhaustive, definition for “Software as a Medical Device” (SaMD) as “software intended to be used for one or more medical purposes that perform these purposes without being part of a hardware medical device.”
In its June 2023 Roadmap, the MHRA announced it is developing guidance to clearly identify SaMD, differentiating it from other tools like wellbeing and lifestyle software products, in vitro diagnostic (IVD) software, and companion diagnostics. The guidance will also cover requirements for ‘in-house’ SaMD used by hospitals, research use only exemptions, custom-made devices, software in kits, software systems, software in procedure packs, software as a service, and accessories to medical devices or IVDs, as well as devices with no medical function[1]. Similar guidance helping to identify regulatory boundaries is available from EU authorities and US FDA.
Regulatory nuances in the EU, UK and USA
The EU has recently published the AI Act that includes regulates both medical devices and other products incorporating AI functions, aiming to ensure consistency and avoid unnecessary administrative burdens for providers of products containing high-risk AI systems. The proposal specifies that “AI systems identified as high-risk should be limited to those that have a significant harmful impact on the health, safety and fundamental rights of persons in the Union”[2]. In the health sector, where stakes are high, sophisticated diagnostic systems and decision-support systems must be reliable and accurate. The extent of adverse impacts on fundamental rights is crucial for classifying an AI system as high-risk[3] .
The forthcoming EU AI Act may therefore increase regulatory scrutiny for AI-based medical devices, potentially bringing certain digital health products under CE marking regulations for the first time under high-risk AI system assessments.
In the USA, a recent Executive Order is expected to influence how the FDA regulates AI/ML in medical devices. The FDA applies a “benefit-risk” framework, requiring demonstration of sensitivity and specificity for diagnostic devices, validation of intended purposes, and ensuring repeatability, reliability, and performance. The FDA also considers the need for adaptive re-training of certain AI/ML systems, introducing pilot processes for pre-authorized software changes without further regulatory assessment[4]. This innovation addresses challenges faced by traditional EU assessment methods.
To keep pace with evolving ML methods, the MHRA in the UK will focus more on guidance than regulation, allowing more frequent updates to keep pace with innovation. Alongside the FDA and Health Canada, the MHRA has outlined 10 guiding principles for Good Machine Learning Practice (GMLP) to ensure safe, effective, and high-quality AI/ML medical devices.
In 2023, the MHRA also updated the Software and AI as a Medical Device Change Programme to ensure clear regulatory requirements and patient protection. This programme builds on broader medical device reforms and innovations, including the Medical Technology Strategy Report (2023) and a £30 million investment in innovative technology for the NHS. The Change Programme aims to make the UK a leader in responsible innovation for medical device software by focusing on safety assurances, clear guidance for manufacturers, and collaboration with partners like NICE, NHS England, and international regulators.
Read more on the subject in our whitepaper “Digital Dilemmas: Regulatory challenges for Artificial Intelligence and Machine Learning in medical devices and digital health products” and make sure to follow us on LinkedIn to not miss any of our blogs.
[1] Gov.co.uk, Software and AI as a Medical Device Change Programme – Roadmap, June 2023
[2] EUR Lex, Proposal for a Regulation laying down harmonised rules on artificial intelligence, (27)
[3] EUR Lex, Proposal for a Regulation laying down harmonised rules on artificial intelligence, (27)
[4] FDA, Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence/Machine Learning (AI/ML)-Enabled Device Software Functions Draft Guidance for Industry and Food and Drug Administration Staff APRIL 2023,
