Having a stroke is devastating. It can leave one partially paralysed, unable to speak and it can kill.
This is why you need to act FAST (face, arms, speech, time). The idea is simple: earlier treatment leads to better outcomes.
Faster — and FAST-er — treatment has a much greater chance of returning patients to normal life than hope and rehab, which is all we used to offer. Recent studies have shown that eStroke, a software package from the Oxford University start-up Brainomix that analyses brain scans and uses artificial intelligence (AI) to make diagnoses, reduced door-in to door-out times for patients by more than an hour. It also tripled the number of stroke patients achieving functional independence from 16 per cent to 48 per cent.
A stroke is caused by a loss of blood supply to part of your brain and can be caused by a thrombosis (blood clot) or the opposite, a haemorrhage. The treatments are opposite too so when a patient has a stroke making the right judgment call is critical to successful treatment: is this a thrombosis that needs a thrombectomy (removal of a blood clot), thrombolysis or blood thinning, or is it due to a haemorrhage — less treatable, but made worse by the wrong treatment? This is where AI can make a world of difference.
e-Stroke is a software package that analyses your brain scan as you emerge from the scanner, to tell your doctor whether it is a blood clot that can be treated by thrombectomy or thrombolysis, or a haemorrhage that needs evacuation or leaving alone.
Advertisement
In the past we relied on radiologists, but they are frequently unavailable straight away because they are so busy with other clinical interventions. e-Stroke enables clinicians to make decisions FAST, to refer immediately and enable treatment to be delivered quicker and more effectively than ever before. At Barts and the Royal London, the use of eStroke increased the number of thrombectomies 2.5-fold in the first year of deployment, and improved outcomes: a clinician-led improvement in patient outcomes that has led to more than 50 publications.
The public might ask: if this is so good, why is it not everywhere? Exactly. The NHS is risk averse, looks at short-term cost, is slow to change and slow to improve —but eventually it will get there.
Professor Kevin Moore is professor of hepatology at the Royal Free London NHS Foundation Trust
