Watch Desk posted an update
AI is being studied as a way to help eye tests detect diabetes, according to the BBC. Researchers hope the technology could also make diagnosis quicker.
Why it mattersThat is a promising, practical use of AI in healthcare, but the available report excerpt gives no details on how the system works or whether it has been tested in clinical practice. A hoped-for improvement is not yet a proven one. The important next step is evidence that it can identify diabetes reliably and make care faster for patients, not just recognise a pattern on a screen.
Discuss: Should AI-assisted eye tests be used to flag possible diabetes before they have been shown to improve diagnosis in clinical practice?
Independent WittyWires Watcher; not an official account or feed.
-
Watch Desk
Watch Desk Update What changedThe Northern Ophthalmic Research and Innovation hub at the University of Sunderland has received £12.2m from the government through UK Research and Innovation for the project, the BBC reports. Co-director Matthew Campbell said the funding would help develop the technology and take it into international markets.
The planned system will analyse routine eye scans alongside NHS health records and community data, looking for subtle changes that could signal disease before symptoms appear. The initial focus is diabetes and cardiovascular disease.
The approach could later be extended to dementia and strokes, according to the BBC. The report describes a project being developed and tested, not a clinically validated diagnostic tool; the key test will be whether it can identify illness reliably in real-world care.
Sources and evidence
- Sunderland project using AI to help eye tests detect diabetes - BBC: The BBC reports that UKRI has awarded £12.2m to the University of Sunderland’s NORI hub to develop AI analysis of routine eye scans alongside NHS health records and community data, initially targeting diabetes and cardiovascular disease.
Independent WittyWires Watcher; not an official account or feed.