MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641126774 A) filed by Srm Institute Of Science And Technology, Ramapuram Campus; and Easwari Engineering College on October 24, 2026, for Artificial Intelligence-Powered Ophthalmic Diagnostic System And Method For Automated Fundus Image Analysis, Biomarker Quantification, And Disease Severity Classification.
Inventors include Stephen Rodrick; and Dr. Madhurima Joardar.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: An artificial intelligence-powered ophthalmic diagnostic system and method (OpticNova) for automated analysis of retinal fundus photographs is disclosed. The system integrates a CLAHE preprocessing pipeline, a convolutional neural network producing a glaucoma probability score with gradient-based visual explainability, and an encoder-decoder segmentation network generating optic disc and optic cup masks. A mathematical biomarker engine computes Cup-to-Disc Ratios in multiple formulations with novel confidence and bootstrap uncertainty scores; ISNT quadrant rim analysis with statistical pairwise comparisons, continuous compliance scoring, and a Fourier asymmetry index; retinal vessel density, tortuosity, and fractal dimension; and macular drusen detection using a weighted multi-color-space enhancement with particle morphometry and spatial density mapping. All biomarkers combine into a composite disease severity score using a calibrated weighted model with Bayesian posterior probability and epistemic uncertainty quantification. A privacy-preserving accessibility module generates auditory diagnostic reports without transmitting raw image data, and a report module produces a structured clinical summary.
Disclaimer: Curated by HT Syndication.
