Researchers at Imperial College London and Imperial College Healthcare NHS Trust have developed the tool.

The Trust includes hospitals like Queen Charlotte’s and Chelsea and Hammersmith.

It’s thought up to 1.2 million people have Type 2 Diabetes but are yet to be diagnosed.

Doctors believe the technology could allow for early detection as much as ten years before patients develop the condition.

The research, funded by the British Heart Foundation has been presented at the American Heart Association’s Scientific Sessions in Chicago.

The team, led by Dr Fu Siong Ng, a consultant cardiologist, and Dr Arunashis Sau, a cardiology specialist registrar at Imperial College Healthcare NHS Trust, developed the AI-ECG Risk Estimation for Diabetes Mellitus (AIRE-DM) tool, using around 1.2 million ECGs from hospital records.

Using data from the UK Biobank, they then validated the AI’s ability to detect subtle changes in routine ECGs that could signify that someone might be at higher risk of type 2 diabetes, years before their blood sugar levels begin to rise.

AIRE-DM accurately predicted future risk in people of various ages, genders, ethnicities and socioeconomic backgrounds about 70% of the time. The researchers suggest it could help spot people who might otherwise not have been identified as likely to develop the condition.

Dr Libor Pastika, clinical research training fellow at Imperial College London, who was also awarded the early career investigator award at the AHA, said:

“AI holds enormous potential to transform care that could lead to substantial improvements in health. By using AI to unlock insights hidden within ECG data, AIRE-DM could be revolutionary in identifying future risk of type 2 diabetes early on.

“By offering a cheap, accessible, non-invasive way to predict type 2 diabetes risk early, AIRE-DM could open up a new window of opportunity for more targeted, preventative care. Supporting people early on to make simple lifestyle changes could help more people avoid type 2 diabetes, and its associated complications.”

When the team incorporated the AI predictions with genetic and clinical information, such as age and blood pressure, it improved the accuracy even further, providing an even clearer picture of risk.

The AI will be piloted in the next year, and the researchers hope it could be rolled out in the NHS in the next few years.