Imperial College London and Edinburgh University have developed an AI software said to read the brain scans of patients who have had a stroke, to understand when the stroke happened and whether it can be successfully treated.
The AI algorithm was trained on a dataset of 800 brain scans where the stroke time was known. Researchers taught it to automatically find the relevant area from the brain scan and then read and analyse the identified lesions to produce an estimated time of when the stroke occurred.
Tests were conducted on 2,000 different patients, including those from Imperial College Healthcare NHS Trust, with researchers reporting on the AI software being “twice as accurate as using a standard visual method”. They also noted the extent of the AI’s capabilities, explaining how it could estimate the biological age of the lesions, helping to determine whether the stroke may be reversible.
Dr Paul Bentley, consultant neurologist, said: “For the majority of strokes caused by a blood clot, if a patient is within 4.5 hours of the stroke happening, he or she is eligible for both medical and surgical treatments. Up to six-hours, the patient is also eligible for a surgical treatment, but after this time point, deciding whether these treatments might be beneficial becomes tricky, as more cases become irreversible. So it’s essential for doctors to know both the initial onset time, as well as whether a stroke could be reversed.”
Dr Bentley added: “Not only is our software twice as accurate at time-reading as current best practice, but it can be fully automated once a stroke becomes visible on a scan.”
AI software solutions in healthcare
Imperial College Healthcare NHS also recently introduced an AI tool to help predict type 2 diabetes up to 10 years in advance, by analysing subtle changes in EGC readings during routine heart scans. the tool is said to be accurate around 70% of the time.
Recently, the Scottish Government gave their annual update on progress toward the NHS Recovery Plan. It highlighted a range of aspects concerning the role of digital innovation in healthcare, including its ability to empower patients, support preventative care, manage demand and more. The update also covered the ongoing work to develop AI policy and guidance for health settings, as well as noting the progress made around developing a digital front door, enabling remote monitoring, and enhancing scheduling.
Seven Manchester trusts are set to utilise AI imaging to help detect diseases such as lung cancer. The project forms part of a wider programme led by the Greater Manchester Cancer Alliance with the tech said to help read chest X-rays by detecting up to 124 findings on chest radiographs and potential lung cancer cases, with the information said to be “relayed to the reporting medical provider in under a minute”.
Read about our other stories on AI in healthcare or check out our upcoming events to get involved in the conversation.
