How 2 London Hospitals are trialling use of AI to detect early signs of type 2 diabetes.
Imperial College and Chelsea and Westminster Hospital NHS Foundation Trusts have begun training an AI system named Aire-DM to analyse patients' ECG heart traces, which is then used to identify subtle early warning signs of type 2 diabetes that are often difficult for doctors to detect.
The success rate is high at around 70% but providing the AI with additional details about background risk factors—such as a patient’s age, sex, and whether they have high blood pressure or are overweight—can enhance its predictive accuracy.
Clinical trials are planned for 2025 with up to 1,000 patients at both hospitals having ECG scans read by the AI system to see if it helps detect and predict diabetes.
The work is being funded by the British Heart Foundation and could be rolled out in the NHS in the next few years.
Leave a comment (all fields required). Please note, we are unable to respond to individual comments posted here.