A study conducted at two eye care centres—Aravind Eye Hospital, Madurai and Sankara Netheralaya, Chennai—and published in JAMA Ophthalmology in March 2020, showed that the artificial intelligence’s (AI) performance exceeded the conventionally used manual grading method, used to identify diabetic retinopathy with a specificity and sensitivity of around 90 per cent. Today, AI has become a valuable extension of the medical field. In the long list of uses of AI, an addition has been made by high-tech screening tool has been developed by Google made by a team of researchers for detecting diabetic retinopathy.
Study
In the study, some 3000 patients with diabetes were screened through a specialised retinal fundus camera to take photos of the eye. In this technique, a coherent beam of light can enter the small gap (pupil) and can take an image in just two-three minutes. The images were fed into the computer and the AI tool screened it for diabetic retinopathy.
In previous study published in 2016, 12,000 images of the retina were taken through AI tool. As per the International Clinical Diabetic Retinopathy scale, the AI tool was taught to grade the severity—none, mild, moderate, severe or proliferative—and give an instant report along with the recommendations.
Since diabetic patients are asymptomatic when the eye is concerned, until the advanced stage when the treatment becomes difficult, it is quite important that diagnosis is made at an early stage. For this, an opportunistic screening has been developed in which doctors place camera at the diabetologists’ clinic and technicians take a picture of the back of the eye (retina) and upload it to the AI tool. In about two minutes, the patient can get their eye test report along with the other regular diabetic test reports. Based on this report, the diabetologist can further refer the patient to an ophthalmologist as per the requirement.
The study found that out of 100 patients screened, about 20 have any level of diabetic retinopathy and only four to five might need intervention