How to detect major complications of diabetes by using AI scanning a patient’s eyes


Researchers from the University of Liverpool and Manchester Metropolitan University modify equipment currently used by high street optometrists to detect diabetic peripheral neuropathy ((file image of a doctor examining a patient's eye)

Artificial intelligence could detect a major complication of diabetes by scanning a patient’s eyes.

Researchers from the University of Liverpool and Manchester Metropolitan University are adapting equipment currently used by high street optometrists to detect diabetic peripheral neuropathy (DPN).

It will work by scanning nerves in the front of the eye rather than the back, with the device’s AI element being able to predict future damage.

Dr. Uazman Alam, from the Institute of Life Course and Medical Sciences at the University of Liverpool, said: ‘What we know from a body of work I’ve been very closely involved with over the past 15 to 20 years is that the nerves at the front of the eye actually reflect nerve damage elsewhere in the body.’

DPN is a major complication of diabetes and the leading cause of limb amputation in diabetic patients.

Researchers from the University of Liverpool and Manchester Metropolitan University modify equipment currently used by high street optometrists to detect diabetic peripheral neuropathy ((file image of a doctor examining a patient’s eye)

It is caused when high blood sugar levels damage the nerves that send messages from the brain and spinal cord to the rest of the body.

The team has been awarded £1.4 million to develop the new machine, which is essentially a redesigned optical coherence tomography (OCT) device, a tool used by optometrists to scan the back of the eye.

The test currently used to detect sensory loss in the limbs of diabetic people is called the 10 gram monofilament.

But the researchers say the “raw” screening, which measures the nerves in a person’s foot, currently misses many people with the condition.

Dr. Alam said: ‘At the moment (patients) are being screened, but the tests we use are not sensitive. We hope this will be a lot more sensitive.

“Instead of having to take measurements on the nerves, we can use the whole picture to detect the nerve damage and actually predict who will develop it.”

In June, a study suggested that more than a billion people worldwide could live with diabetes in the coming decades.

The paper, published by the journal The Lancet Diabetes and Endocrinology, said some 1.3 billion people will have diabetes by 2050 – more than double the 529 million cases in 2021.

DPN is a major complication of diabetes and the leading cause of limb amputation in diabetic patients (file image of a doctor examining a patient's eye)

DPN is a major complication of diabetes and the leading cause of limb amputation in diabetic patients (file image of a doctor examining a patient’s eye)

It is hoped that the study will be completed in 2027 and eventually result in a pilot clinical validation trial in healthy and diabetic volunteers at Aintree University Hospital in Liverpool.

Dr. Alam predicts that AI will “one day become an important facet of all healthcare systems,” but “needs further development” before widespread adoption.

“I think we need to remember that AI isn’t just the images we’re talking about, it can also be data,” he said.

‘It’s here to stay. We have to develop it in an ethical way.

“I think it’s important and I think it should probably be taught in medical schools as well. It will settle in healthcare.’