Spotting sicknesses and saving lives: Durham’s new AI tools for hospitals

Life-saving technology from Durham University’s Computer Science department


Durham University’s Computer Science department has created AI technology to help doctors and nurses identify critically ill patients earlier on.

The invention, outlined in a paper published in BMJ Health and Care Informatics, could directly save lives by enabling faster treatment.

The AI tools will also aid communication between hospitals and patients who are leaving.

The technology was developed by Durham University in partnership with The Northern Care Alliance NHS Foundation Trust and the University of Greater Manchester. The National Institute for Health and Care Research (NIHR) and the Medical Research Council (MRC) also provided support.

The project won a Best Presentation Award at the Society for Acute Medicine’s International Conference. It was in an NIHR policy briefing to the Department of Health and Social Care.

Currently hospitals often use the NEWS2 system, which relies on limited data interpreted by humans. However, the rise of electronic health records allows an AI tool developed by Professor Noura Al Moubayed and Dr Matthew Watson to examine a patient’s lifelong health record. This will produce a more accurate and personalised risk score due to a wider array of information. Hospital staff can then identify which patients are most likely to become critically ill quickly.

This model has been tested on over 170,000 admissions at Salford Royal Hospital, and it performed far better than classic NEWS2 systems. According to a piece produced for Durham University’s news page, it correctly identified 92 per cent of patients whose condition later got worse. This is a step up from NEWS2’s 13 per cent. A full report on the testing of this technology is on the BMJ Journal’s website.

A positive use for AI

In a world of increasingly worrying uses for AI, this comes as good news, inspiring hope for further constructive and ethical uses for AI in treatment that can save lives. Being merely a tool that evaluates high and low levels of risk, it is relatively safe for practical use. This means its implementation in hospitals will likely bring further positive results – but lives aren’t at stake if it doesn’t.

For more of the latest news, guides, gossip, and memes, follow The Durham Tab on InstagramTikTok, and Facebook.