MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641050499 A) filed by Kalasalingam Academy Of Research And Education on April 21, 2026, for Texture-Enhanced Deep Learning For Multi-Stage Glaucoma Classification.
Inventors include Dr. Kalpana Murugan; G. Siddiq; M. Tharun; and Y. Madhu Venkat Vardhan.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: Title: TEXTURE-ENHANCED DEEP LEARNING FOR MULTI-STAGE GLAUCOMA CLASSIFICATION The titled invention discloses a texture-enhanced deep learning for multi-stage glaucoma classification, a software-based system designed for early detection and multi-stage classification of glaucoma using retinal fundus images. The system comprises an input dataset module (1) for acquiring retinal images, followed by a pre- 10 processing module (2) that applies median filtering to remove noise while preserving structural details. An edge detection module (3) identifies intensity variations to locate structural boundaries, which are further processed by a Fuzzy C-Means segmentation module (4) to isolate the optic disc and optic cup regions. A feature extraction module (5) utilizing Gray-Level Co-occurrence Matrix (GLCM) computes texture parameters 15 including contrast, correlation, energy, and homogeneity. The extracted features are processed by a Deep Convolutional Neural Network (DCNN) classifier module (6) to categorize the condition as benign or glaucomatous with multi-stage severity grading. The output module (7) presents the diagnostic results along with performance metrics. The invention provides a cost-effective, reliable, and fully software-implemented 20 solution suitable for large-scale glaucoma screening and early diagnosis.
Disclaimer: Curated by HT Syndication.
