MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202631041845 A) filed by National Institute Of Technology, Patna, Bihar, on April 1, for 'system and method for automatic detection of ocular disease using convolutional kolmogorov-arnold network.'
Inventor(s) include Preity; Bhandari, Ashish Kumar; and Shahnawazuddin, Syed.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to a system(100) and method for automatic detection and classification of ocular diseases from retinal fundus images using a deep learning based architecture. The system includes an image acquisition module(101), an image preprocessing module(102) configured to perform normalization, resizing, and train-test splitting, and a deep learning model(103) comprising convolutional Kolmogorov-Arnold Network (ConvKAN) layers and Kolmogorov-Arnold Network (KAN) linear layers. The ConvKAN layers utilize spline-based nonlinear kernel functions to extract spatial features from retinal fundus images, while the KAN linear layers perform disease classification or disease severity grading based on the extracted features. An output module(105) generates a disease detection output indicative of one or more ocular diseases or disease severity levels. The invention is capable of performing binary classification, multi-class classification, and severity grading of ocular diseases including cataract, diabetic retinopathy, and glaucoma, while achieving parameter-efficient feature extraction and improved performance."
Disclaimer: Curated by HT Syndication.
