Hyderabad doctors develop AI-smartphone tool with 92% accuracy for glaucoma detection

Hyderabad’s LV Prasad Eye Institute has developed an AI-powered smartphone tool that detects glaucoma with over 92% accuracy, including 86.9% in early-stage cases. The low-cost, offline tool could revolutionize early diagnosis, particularly in rural and underserved communities.

Hyderabad: Eye specialists from Hyderabad have developed, tested and validated an AI tool that works through a Smartphone camera to detect glaucoma in patients with an overall diagnostic efficiency of 92.02 percent. The new AI-Smartphone tool also has a remarkable 86.9 percent accuracy in detecting early-stage glaucoma among patients.

Published in the prestigious PLOS One (July 2025), the study by eye specialists from the Hyderabad-based LV Prasad Eye Institute (LVPEI) reported that the AI system has achieved an overall accuracy of 92.02 percent, with a specificity of 94.12 percent for identifying cases that require referral to a higher health care facility.

---

The breakthrough has the potential to dramatically improve the early detection of glaucoma, a group of eye diseases that damage the optic nerve. It is particularly beneficial for high-risk and underserved communities.

The study included 213 participants, carefully chosen to represent high-risk groups of patients. For patients above 65-years and older, the AI tool showed a remarkable sensitivity, which is the ability to correctly identify the disease, at 90.57 percent.

The patients in the study included a subset with a history of diabetes. The AI-tool managed to correctly identify 42 patients of the 45 glaucoma patients, which highlights its ability to screen patients with co-morbid conditions, the researchers said.
In the research paper, the LVPEI scientists said that the AI-tool could be a game changer for patients because it is low-cost, portable and has the ability to function without internet connection, ideal for utilizing in rural areas.

“Our study presents real-world evidence supporting an AI-driven method backed by robust data, poised to address this pressing global public health challenge, particularly in resource-limited settings. The AI-based offline tool integrated on a Smartphone camera showed a promising performance in detecting referral-warranted glaucoma compared to a glaucoma specialist’s diagnosis. The AI showed higher accuracy in detecting advanced glaucoma followed by moderate and early glaucoma. It has the potential to enhance screening efforts, serving as a beacon of hope in combating this pervasive global issue,” the LVPEI researchers concluded.