MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202641111384 A) filed by Srm Institute Of Science And Technology Tiruchirappalli on September 17, 2026, for A Deep Learning Framework For Diabetic Retinopathy Detection And Lesion Localization Using B- Lstm–cnn Architecture.
Inventors include Priyadharshini K S; Nivetha G; and Dr. Devisri B.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: Learning framework for real-time, remote screening. The system uses a camera module based on the ESP32 microcontroller to acquire, preprocess, and control the retinal images. The images are analyses using a hybrid deep learning model that uses shared feature extraction and bidirectional learning to learn both structural and contextual retinal features. The adaptation of the features optimally reduces computational requirements while maintaining diagnostic accuracy. The unified model performs screening for the severity and localization of retinopathy lesions, allowing the system to operate on low-power edge devices. Through linked platforms, the diagnostic outputs can be transmitted to doctors, enabling remote screenings and tele ophthalmology. This invention provides a scalable and cost-effective approach to early diabetic retinopathy detection in underserved communities.
Disclaimer: Curated by HT Syndication.
