UK government to invest in creating artificial patients for clinical trials

Healthcare researchers in the UK could see artificial patients take part in clinical trials as the UK government invests close to £1m on projects, including AI in healthcare.

The Medicines and Healthcare products Regulatory Agency (MHRA) has received nearly £970,000 from the Department for Business, Energy and Industrial Strategy (BEIS) – both a part of the UK government – to launch projects aiming to “unlock digital, data and scientific regulatory innovation”.

The first AI-related project aims to tackle the challenges of finding control groups for clinical trials, through the development of alternative synthetic datasets.

As it stands, patients run the risk of being randomly assigned to control groups and therefore are denied access to treatments that could improve their symptoms or extend their life, and clinical trials for rare diseases can often find it difficult to recruit enough patients for their trials altogether.

According to the government, the MHRA is on its way to creating artificial patients, who have similar health information to real patients, that could take part in small trials and “better reflect groups in society that are less well represented in clinical trials,” such as different age groups and ethnicities.

Dr Puja Myles, director of the Clinical Practice Research Datalink (CPRD) at the MHRA said: “This funding will allow us to build upon the innovative work we have developed here at CPRD, both before and during the pandemic, creating synthetic datasets that mimic real patient data for use in clinical trials.”

“This provides researchers with high-fidelity data that replicates the complex clinical relationships in real primary care data while protecting patient privacy, as they are wholly synthetic.”

Another project within the investment will look into the issue of how to safely introduce complex ‘black-box’ AI products into clinical settings.

Black-box AI describes an AI product whose systems inputs and operations are not visible to the user.

Nearly £168,000 has been awarded to this project with the hopes to produce a workable methodology that supports the regulation of both transparent and complex AI models.

“Patient safety is at the heart of everything we do, and we’re determined to be well prepared to get AI technology to the bedside safely, effectively and with confidence in this assurance,” said Johan Ordish, head of software and AI at the MHRA.