News

Milton Keynes University Hospital implements AI-powered software to help stroke patients

An AI-powered software known as ‘e-Stroke’ has been implemented in Milton Keynes to help care for stroke patients.

The software analyses computerised tomography (CT) scans, identifying any areas of the brain that may be damaged and automatically flagging them to doctors.

It is hoped that this will aid doctors in making quicker clinical decisions, saving time when a patient is suffering from a stroke and providing patients with an improved chance of recovery.

Dr Janet Costa, Consultant Stroke Physician from Milton Keynes University Hospital, has described how she was able to send images of a blood clot in a patient’s brain instantly to a specialist colleague, Dr Phil Mathieson from Oxford University Hospitals, resulting in a speedy transfer for the patient for the necessary surgery.

Dr Mathieson said: “It’s certainly been a game-changer for me in terms of the speed at which we are able to look at images from the region and make transfer decisions in a much, much quicker timescale than we have previously been able to.”

Nina Roberts, Stroke Advance Nurse Practitioner at Milton Keynes University Hospital, said: “The Brainomix software is a fantastic tool to help us in the diagnosis and treatment of a patient suffering from a stroke. Every minute counts when it comes to treating a stroke. The AI software helps us speed up the treatment of a patient by being able to view CT scans on a smartphone and get expert advice quickly from other stroke specialists.”

Dr George Harston, Chief Medical and Innovation Officer at Brainomix, highlighted how e-Stroke “really focuses on the simple imaging, the imaging that is available in every hospital” in order to provide opportunities to “improve stroke care for all clinicians and for all patients, whichever hospital they go to.”

Earlier in the month HTN reported on an AI tool to support cardiovascular care developed in Sheffield and a deployment engine for AI that has been shared as open source.