Skip to main content

Research and Projects

SAMueL – Stroke Audit Machine Learning

Who is involved?

Theme: PenCHORD
Status: Live



Stroke is a leading cause of death and disability with over 85,000 people hospitalised in the UK each year. One way of treating stroke and preventing disability is to give a patient medication to break down blood clots. This is called thrombolysis. 

However, thrombolysis is not suitable for all patients and can be risky. For thrombolysis to be useful it needs to be given as soon after the stroke as possible. Use of thrombolysis varies hugely, even for patients with similar treatment pathways and with similar characteristics; some hospitals rarely use it while some use it in a quarter of stroke patients. The speed of giving thrombolysis also varies, with some hospitals taking an average of 90 minutes and others taking less than 40 minutes to administer the drug.

Project Aims

Using modern computer science techniques of clinical pathway modelling and machine learning, we will aim to find out why there is so much variability in the use of thrombolysis. This will help hospitals understand what they can do to optimise its use.

Based on the decisions made by highly qualified qualified stroke experts we will build a tool for assisting doctors to review of their use of thrombolysis. This will be particualarly useful to hospitals without sufficient funding to employ a team of expert stroke physicians.

Project Activity

Study Design: The research team will use a state-of-the-art computer modelling technique -  pathway modelling -  to better understand what causes variation in care across the UK. This approach replicates, in a computer model, the flow of patients through the first few hours of stroke care, mimicking the same processes and timings that the stroke unit currently provides. This allows us to look at the effect of changing key aspects of patient flows in a controlled, modelled environment, without affecting real patients. A second technique called ‘machine learning’ enables us to teach a computer the likely decision made in any hospital given any particular patient.

Both approaches allow us to ask 'what if?' questions, such as 'what if a hospital improved diagnosis of patients by asking more questions, but by doing so extended patients’ waiting times for scans?'. With machine learning we can ask ‘what if the decisions at all hospitals were similar to hospitals that are considered centres of excellence for stroke care?’. By asking these types of questions we can identify changes at each hospital of most benefit to patients. Both techniques have been piloted across seven hospitals.

We would now like to test and refine these methods across all stroke units in England. A researcher will interview doctors to understand their attitudes to thrombolysis, and how the results from the modelling work can best be presented to them in a way that will influence more consistent stroke care across the UK.

Anticipated Outputs

We are conducting this work with the National Stroke Audit, hosted by the Royal College of Physicians. Our aim is to build these new advanced analytic tools into the quarterly stroke audit, helping hospitals understand whether their use and speed of thrombolysis is different from that expected for their patient population, and what changes would most improve performance (if needed).

Next Steps

Regional pilot work has been completed, and the the project is anticipating supply of national data in 3Q 2019. A pilot national model is expected to be complete in 4Q 2019.




Julia Frost (Third Gap)
Richard Everson (Computer Science)
Zhivko Zhelev (Technology Assessment Group)