New AI initiative in Devon aims to improve stroke treatment outcomes by analysing extensive medical data

A pioneering research initiative led by the Royal Devon University NHS Foundation Trust, in collaboration with the NIHR Applied Research Collaboration South West Peninsula (PenARC) and the University of Exeter, is harnessing artificial intelligence (AI) and machine learning to transform stroke treatment across the United Kingdom. This innovative project aims to enhance outcomes for people affected by stroke by improving how clot-dissolving treatments are delivered.

Personalising Stroke Care Through AI

The research focuses on the use of thrombolysis—a time-sensitive treatment that involves administering medication to break down blood clots in the brain, thereby limiting the extent of disability following a stroke. While thrombolysis has the potential to significantly improve recovery outcomes, it is only appropriate for certain individuals and must be administered rapidly after the onset of stroke symptoms.

Currently, thrombolysis is provided to approximately 11% of people experiencing a stroke in the UK. In the South West of England alone, more than 1,000 individuals receive this treatment each year. However, national data reveals considerable variation in both the speed and frequency with which thrombolysis is delivered in hospitals, contributing to inequities in care.

The newly launched research initiative—entitled SAMueL 2 (Stroke Audit Machine Learning 2)—seeks to address this variation through the integration of AI into a national stroke audit, the first initiative of its kind globally. By analysing extensive datasets, the team has developed a decision-support tool that helps clinicians identify which people are most likely to benefit from thrombolysis, based on their individual clinical characteristics. This tool is designed to ensure that treatment is delivered both more quickly and more equitably.

Harnessing National Stroke Data

Building on earlier research, the SAMueL 2 team employed sophisticated computer modelling to investigate why thrombolysis rates differ so significantly between hospitals. The models also helped to predict outcomes for individuals who received thrombolysis compared to those who did not, shedding light on the factors most strongly associated with positive recovery outcomes.

Professor Martin James, Consultant Stroke Physician and Honorary Clinical Professor at the University of Exeter Medical School, explained:

“SAMueL analysis includes nationwide data from a quarter of a million stroke cases, and by using this data, we can provide each hospital with a tailored target for thrombolysis. When teams have used this as a benchmark, they’ve been able to treat more patients, more effectively.

Stroke has a life-changing impact, so it’s inspiring to see how research like this can lead to more personalised, faster treatment and better outcomes for patients and their families.”

The project represents a significant step forward in the use of AI in clinical practice, setting a precedent for how data-driven approaches can inform more nuanced and equitable healthcare delivery. By drawing on patient-level data from across the country, the tool enables NHS trusts to benchmark their performance, optimise treatment pathways, and ultimately improve care for those affected by stroke.

Supporting National Health Goals

This research is part of a wider programme funded by the National Institute for Health and Care Research (NIHR) and aligns with the government’s ambition to increase equitable access to thrombolysis. It underscores the critical role of personalised, data-informed medicine in addressing one of the UK’s leading causes of death and long-term disability.

By combining clinical insight with advanced analytical tools, the SAMueL 2 initiative is not only enhancing individual outcomes but also shaping the future of stroke care at a national level.