MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621048036 A) filed by Moresh Madhukar Mukhedkar; Prashant Kumar; Lokesh Kumawat; Prateek Sarkar; Rahul Patel; Shakil Tamboli; Vivek Patil; and Shweta Koparde on April 15, 2026, for Medipredict: A Deep Learning-Based System For Automated Multi-Class Eye Disease Detection.
Inventors include Moresh Madhukar Mukhedkar; Prashant Kumar; Lokesh Kumawat; Prateek Sarkar; Rahul Patel; Shakil Tamboli; Vivek Patil; and Shweta Koparde.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The present invention relates to an artificial intelligence-based system and method for automated detection of eye diseases using ocular images. The system is designed to assist in early screening by analyzing retinal and external eye images through deep learning techniques. The invention includes an image preprocessing module, an input validation module, and a multi-class disease classification module. A distinguishing feature of the system is an Eye vs Non-Eye classifier that verifies whether the input image contains a valid eye region, thereby reducing incorrect predictions caused by irrelevant inputs. Upon validation, the image is processed using a convolutional neural network-based model to classify multiple eye diseases and generate prediction probabilities. The system further provides a confidence score corresponding to the predicted class. In addition, an interpretability mechanism highlights important regions of the image contributing to the prediction. The system is deployable as a web-based application, enabling real-time and accessible eye disease screening.
Disclaimer: Curated by HT Syndication.
