AI project eyes early stage diabetic retinopathy


In a first for Mumbai, early stage diabetic retinopathy has been detected using artificial intelligence (AI) at civic-run dispensaries. The unique project is being implemented by the Aditya Jyot Foundation for Twinkling Little Eyes (AJFTLE) and, in a span of eight months, nearly 1,300 diabetes patients have been screened on a retinal imaging device attached to a smartphone.

Commenced in August 2018, the Foundation has screened patients in 18 civic-run dispensaries across the city. Till March this year, 92 patients were diagnosed with early stage diabetic retinopathy and referred to municipal hospitals or the Aditya Jyot Eye Hospital for further treatment.

“Diabetic retinopathy is tricky because there are no early signs. That’s why screening is the only way to detect the complication early and prevent diabetic blindness,” said eye surgeon Dr. S Natarajan, also managing trustee of AJFTLE.

Common cause

Diabetic retinopathy is the commonest diabetic eye disease; it damages blood vessels in light-sensitive tissue at the back of the retina.

While some patients may have symptoms like blurred vision or impaired colour vision, it’s a common cause of blindness in the diabetic population. A robust screening programme is thus the need of the hour.

As a part of the AI project, technicians from the Aditya Jyot Foundation visit civic dispensaries along with Remedio Fundus on Phone, a portable device attached to a smartphone equipped with retinal imaging.

After the patient’s eye images are clicked, the AI on the device screens them for signs of diabetic retinopathy, and prompts technicians on whether they should be referred to a hospital or not. If the image is unclear, the device also prompts a retake of the picture.

Besides diabetic retinopathy, 97 patients, who had been referred, were diagnosed with cataract, and 69 had other eye diseases, including retinitis pigmentosa, age-related macular degeneration and glaucoma, of which they were unaware.

Not the last word

“The AI diagnosis cannot be termed as the last word. Thus, the diagnosis is verified by an ophthalmologist when the patient is referred further. But this is an excellent tool to indicate the risk of retinopathy in patients. Also, the added advantage is that the device does not depend on [an Internet] network and is completely offline,” said Dr. Radhika Krishnan, AJFTLE’ chief executive officer, adding the initiative was in the process of expanding to 60 civic dispensaries in Mumbai.

A growing challenge

The World Health Organisation (WHO) states that diabetes is a growing challenge in India, with an estimated 8.7% diabetic population in the age group of 20 to 70 years. “There are an estimated 73 million diabetics in the country. Of these, nearly 25% are at the risk of diabetic retinopathy,” said Dr. Natarajan.

Rural areas

Dr. Natarajan is certain that the screening programme can be replicated anywhere in India, especially in rural areas. “Unskilled or semi-skilled people can be trained to conduct the screening, which will go a long way in preventing blindness,” he said.

Leave a Comment

Your email address will not be published.

You may also like

Hot News