Manchester-led team maps path for virtual patients in device regulation
A consortium led by the University of Manchester has published a framework setting out how computer-generated "virtual patients" could be used alongside conventional clinical testing to accelerate and improve the safety assessment of medical devices. The work, produced with the Medicines and Healthcare products Regulatory Agency (MHRA), industry experts and academics across a network of UK universities, appears in the journal Device on 5 October 2026.
The framework addresses a longstanding gap in regulatory science. Existing standards and guidance cover some aspects of computational modelling, but none provides a sufficiently comprehensive process for judging whether a given computer model is credible enough to inform a specific regulatory decision. That ambiguity has slowed adoption of so-called in-silico trials worldwide, even as device complexity has increased.
The framework in practice
The researchers' approach centres on three steps: identifying the most significant potential patient harms for a given device, assessing whether computer modelling can meaningfully evaluate those harms, and calibrating the level of scrutiny applied to models in proportion to the consequences of a regulatory error. The process is designed to give manufacturers a clear pathway to generating digital evidence that complements rather than replaces laboratory experiments, animal studies and human clinical data.
To test the framework, the team applied it to a hypothetical redesign of a transcatheter aortic valve implantation (TAVI) device. TAVI is a well-established procedure for replacing a diseased aortic valve without open-heart surgery, but the implanted valve must fit precisely to avoid pressure on surrounding cardiac tissue and prevent paravalvular leakage. The researchers demonstrated that simulation can examine variables, such as tissue contact pressure, micro-gaps around the valve and atypical blood-flow patterns, that are difficult or impossible to measure directly in a living patient.
Professor Alex Frangi, lead scientist and Executive Director of the UK Centre of Excellence on In-Silico Regulatory Science and Innovation (UK CEiRSI), said the tools used to evaluate increasingly complex medical devices had not always kept pace with that complexity. "Through UK CEiRSI, we are working to provide a practical route for using advanced computer simulations in a way that regulators, manufacturers and clinicians can all trust," he said.
MHRA co-author Mark Grumbridge, Head of Clinical Investigations at the agency, noted that larger and more diverse virtual patient groups could ultimately reveal risks missed by conventional studies, while reducing the number of people exposed to unproven devices during testing.
Regulatory and market context
The publication arrives at a moment of renewed regulatory interest in computational evidence on both sides of the Atlantic. The US Food and Drug Administration has been developing its own guidance on in-silico methods under the Digital Health Centre of Excellence, and the European Medicines Agency has explored similar territory for drug modelling. Harmonisation across the UK, EU and US regulatory systems is an explicit aim of the UK CEiRSI framework, which could have significant commercial implications: a common credibility standard would reduce the duplication of evidence packages that manufacturers currently face when filing in multiple jurisdictions.
UK CEiRSI is funded by UK Research and Innovation through Innovate UK and the Medical Research Council, and is sponsored by the MHRA. Its consortium spans eight leading UK universities and more than 85 national and international organisations. The next phase of work, the team says, will focus on expanding virtual patient populations to reflect real-world diversity more faithfully, an ambition that will require validated methods for generating representative synthetic cohorts at scale.