Doctors at LV Prasad Eye Institute have validated an AI tool for glaucoma detection using smartphone cameras. The tool, tested on 213 patients, demonstrated a 92% efficiency rate. It uses a portable, non-mydriatic fundus camera to capture detailed images of the back of the eye.
Compared to diagnoses from glaucoma specialists, the AI tool showed high accuracy, with 91.36% sensitivity and 94.12% specificity in identifying patients needing referral. For diabetic patients, the tool identified 42 out of 45 glaucoma cases. In patients over 65, it achieved 90.57% sensitivity and 100% specificity. The AI is low-cost, portable and functions offline, making it suitable for rural areas.
