Doctors at the LV Prasad Eye Institute (LVPEI) have successfully tested and validated an artificial intelligence (AI) tool that detects glaucoma using smartphone cameras. Published in the online journal PLOS, the study reported that the tool achieved nearly 92% efficiency.
Glaucoma: The Silent Threat
Glaucoma is a group of eye disorders that damage the optic nerve, which carries visual signals from the eye to the brain. Often called the “sneak thief of sight,” the disease progresses silently, with most patients showing no symptoms until it reaches advanced, vision-threatening stages. Therefore, early detection becomes both critical and challenging.
How the Study Was Conducted
LVPEI researchers collaborated with Remidio Innovative Solutions to test an offline AI tool integrated with a portable, non-mydriatic fundus camera. This device, powered by smartphone technology, captures high-resolution images of the back of the eye.
As reported by TOI, the study screened 213 patients (418 eyes) and classified them as glaucoma-free, glaucoma suspects, or confirmed glaucoma cases. Confirmed cases were further categorised as early, moderate, or advanced.
High Accuracy in Screening
When compared with diagnoses made by glaucoma specialists, the AI tool demonstrated impressive accuracy. It achieved a sensitivity of 91.36% and a specificity of 94.12% in detecting patients who required referral.
To clarify, sensitivity measures the tool’s ability to correctly identify individuals with glaucoma, while specificity reflects its ability to correctly identify those without the disease. Among diabetic patients—a high-risk group—the tool successfully identified 42 out of 45 definite glaucoma cases. For patients over 65 years, it recorded 90.57% sensitivity and 100% specificity.
Wider Implications and Public Health Efforts
Globally, glaucoma affects about 1 in 30,000 people, with higher risks observed in African and Asian populations. Recognising this, Telangana recently opened its first glaucoma centre at a public health centre (PHC) in Nalgonda. Authorities now plan to replicate this model across the state to improve early detection and treatment access.