Pune: A smartphone-based artificial intelligence system has shown high accuracy in detecting major eye diseases including diabetic retinopathy, age-related macular degeneration (AMD) and glaucoma, according to a new study by NIO Super Specialty Hospital.
The study titled ‘Smartphone-based offline AI for multi-disease retinal screening: Real-world accuracy’, evaluated the diagnostic accuracy of an offline AI system called Medios-AI (MAI), which is integrated into a smartphone-based fundus camera.
Researchers found that the MAI algorithm could effectively screen the images captured by the fundus camera for diabetic retinopathy, glaucoma and AMD.
It generated disease specific reports with sensitivity of 99.3% and specificity of 95.7%, as per the study published in the European Journal of Ophthalmology.
The platform works entirely offline, unlike other AI systems that need internet connectivity, making it suitable for outreach camps and teleophthalmology programmes in underserved areas.
The MAI system has received regulatory approval from the Central Drugs Standard Control Organisation (CDSCO) in India for clinical use and carries European Class II medical device certification.
The study was co-authored by Dr. Aditya Kelkar, Dr. Jai Kelkar, Dr. Yash Garg, Dr. Harsh Jain, and Dr. Sabyasachi Sengupta from the NIO Super Specialty Hospital.
Dr Aditya Kelkar, Director, NIO Super Specialty Hospital said “At present, the system is validated for detecting Diabetic Retinopathy, Glaucoma, and AMD. While AI technology helps for detecting additional retinal diseases, each new condition requires independent algorithm development validation, and regulatory approval before clinical use. Thereby, there is a scope for the expansion of this technology to be able to detect more retinal conditions”.
