Wellcome funding puts digital technology at the heart of medical research

Spotting brain injuries from babies’ brainwaves and predicting the best way to tailor stroke care to individual patients are two examples of how digital technology is transforming health. The projects are funded through our Innovator Awards.

Baby in a cot
Professor Geraldine Boylan and her team at University College Cork are using machine learning to identify newborn babies who need treatment quickly.

Earlier this month, we launched £20m of funding through the Innovator Awards to specifically support researchers who want to create digital healthcare innovations that could have a significant impact on human health.

As Wellcome’s Director of Innovation Stephen Caddick explains: "Digital innovation offers a unique opportunity to improve human health and help people living in poverty. We are offering funding to innovators who want to create and develop new digital interventions at speed and scale, and who have the ambition to improve the lives of millions of people."

Recognising brain injury from babies’ brainwaves

Time is critical for doctors treating newborn babies with suspected brain injuries. The sooner treatments, like whole body cooling, can be used, the better the outcomes are likely to be.

Doctors rely on the results of EEG brain monitoring to give them crucial information. But these brain scans are so detailed they can only be interpreted by an expert, and there aren’t enough experts to be at every cot side. 

Wellcome-funded researchers at the INFANT Centre at University College Cork are developing the first ‘smart’ system to recognise patterns in electrical brain activity, which will help to identify babies who need treatment quickly.

Professor Geraldine Boylan and her team are training computers to learn EEG patterns and how they relate to the extent of the brain injury. If the computer can identify the warning signs, this could help more babies to survive, and cut the risk of permanent disabilities such as epilepsy, cerebral palsy or learning difficulties. 

This is just the tip of the iceberg for digital technology in healthcare. As Professor Boylan says: "Machine learning has given us a new tool to improve clinical decision making. These new technologies can transform how we care for our patients."

Watch Professor Geraldine Boylan explain how her team are training computers to learn EEG patterns to help babies at risk of brain injury.

Tailoring stroke care to individual patients

To give patients who have had a stroke the best chance of recovery, speed is crucial. A half an hour delay in treating a stroke patient may equate to weeks of rehabilitation. But surprisingly, little is known about how best to tailor stroke care to individual patients. The brain is an incredibly complex organ and to fully understand conditions like stroke, researchers need to gather vast quantities of data.

Step forward artificial intelligence and machine learning. At University College London, Dr Parashkev Nachev and his colleagues are developing a new system that will help to determine the best treatment for individual stroke patients. 

At University College London Hospitals, Dr Nachev has built a supercomputer that he is now training with the anonymised data from thousands of stroke patients. The computer will learn patterns in the outcomes for different types of strokes and the impact different treatment decisions can have. It will be able to predict the best clinical decisions to make for each patient, speeding up treatment and hopefully improving recovery.

Parallel systems will be developed at King’s Health Partner hospitals, in collaboration with clinicians and machine learning engineers at King’s College London. 

Dr Nachev says: "Improvement in stroke outcomes has lagged behind many other similar conditions, held back by the complexity of the brain. I believe machine learning will be at the heart of how medicine will be done in the future. But it needs support and funding to be developed into systems that can truly transform care."

Related links