Adsilico, a University of Leeds spinout using AI and simulation to assess medical device performance in virtual patient populations, has raised £10m in Series A funding led by PXN Ventures, with participation from existing investors Northern Gritstone and Parkwalk. Funding supports expansion of adsilico’s heart models and extension into devices for the brain, as well as growing the team from 14 to 34 people.
Heart valves, stents and other implants typically reach the market after clinical trials involving a few hundred patients. Recruiting trials that reflect everyone who will receive a device is slow and costly, and some groups are harder to reach, so participants tend to be disproportionately men. A study in JAMA Internal Medicine of 195 trials of high-risk implants found that women made up just 33% of participants, falling to 29% in cardiovascular trials. That leaves manufacturers and clinicians with limited evidence on how a device will behave in women, older patients or people with smaller or different anatomy until devices are in wider use.
adsilico’s Virtual Chimera technology uses generative AI to build realistic virtual hearts and blood vessels from anonymised clinical scans. Under clinical oversight, adsilico then runs physics-based simulations of how a device behaves inside each model across a synthetic population that reflects variation in age, sex, ethnicity, body size and disease. Insights can show, for example, how a device performs in women aged 55 to 75 of South Asian heritage before it reaches a real patient, helping to close diversity gaps in testing where women, older patients and people with smaller or different anatomy are often under-represented.
Medical device manufacturers can use these simulations to inform strategic decisions such as which sizes of a device to make for intended patients, or to identify target patients who may be at higher risk before committing to further development or clinical trial recruitment. Research for the US Department of Health and Human Services published in JAMA Network Open estimates that 13.7% of complex devices entering non-clinical testing go on to reach approval. adsilico adds a new dimension of evidence alongside laboratory, animal and clinical studies so problems surface earlier, later studies can be better targeted, and human-relevant evidence is available before animal studies begin.
Regulators in the United Kingdom and United States are also pushing for change. The United Kingdom’s independent review on equity in medical devices found evidence that some groups may receive sub-optimal treatment because of bias in the design and use of devices. In the United States, the Food and Drug Administration has published guidance on using computer modelling as evidence in medical device submissions and has updated regulations for drugs and biologics to recognise computer models among alternatives to animal studies. adsilico has already completed projects with device makers in heart valve and neurovascular devices, and cardiac work will be presented at TCT, a global meeting for innovations in structural heart therapies. adsilico was founded in 2024 by Sheena Macpherson, Nishant Ravikumar and Zeike Taylor and is based at Nexus, the University of Leeds’ innovation hub.
Patients miss out on the chance of a life-saving device, or face uncertainty about whether it will work for them, because people like them weren’t represented when it was designed and tested. We give medical device manufacturers early insight into the impact of anatomical diversity when it can still shape the design and help avoid costly mistakes or harm to patients. This investment enables us to grow our team and extend our reach beyond virtual hearts into new therapy areas. Our mission is for our virtual populations to reflect real-world diversity, helping make implantable device testing faster, safer and more equitable for the patients who need it. The UK has world-class clinical and research talent, and from our base in the North with backing from UK investment, we set the global standard for virtual patients within a new dimension of in silico generated evidence that will accelerate the launch of safer devices to those that need them most.
adsilico represents the level of groundbreaking innovation and deep thinking being spun out of universities across the North of the UK currently. What Sheena and the team have built has the potential to have a significant global impact for medical device manufacturers, the pre-clinical development process and, therefore most important of all, patients. PXN is excited to be joining at such a vital time for the business, supporting the team as it scales and sets new standards for in silico testing.
Medical device manufacturers need to know that their products will work for the people who receive them, including populations that have been traditionally underserved. adsilico's use of generative AI to build realistic virtual hearts and blood vessels provides vital information to help understand how these products will work much earlier, spotting potential problems before committing to costly human trials. Sheena and the team have come a long way since spinning out of the University of Leeds, and we’re pleased to support them in this next phase of growth.
adsilico is exactly the kind of science we exist to back: deep, hard-won intellectual property out of the UK's universities, with a real chance to change how medical devices reach patients. We've worked with Sheena and the team since the company's earliest days. Their Virtual Chimera technology tackles a problem that has held cardiovascular research back for years: a lack of diverse, representative data in clinical trials. This latest round of funding gives the team the capital to turn that technology into partnerships that put it to work for patients at scale, and we're proud to keep supporting them on that journey.
