MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122954 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 5, 2025, for 'explainable ai system for medical image diagnosis with gradient-weighted class activation mapping.'
Inventor(s) include Dr. Prakash G; and Risa Deshwal.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure provides a system for medical image diagnosis with explainable artificial intelligence capabilities. The system includes a preprocessing module configured to receive and standardize input images, and a convolutional neural network classifier trained to classify the standardized images into categories. A gradient-weighted class activation mapping module computes gradients of predicted class scores with respect to feature maps of a convolutional layer of the classifier, and generates visual heatmaps by weighting the feature maps according to the computed gradients to highlight regions in the input images that contribute to classification decisions. A visualization module overlays the visual heatmaps onto the input images, and a user interface module displays the overlaid visual heatmaps. The system enables transparent medical diagnosis by combining accurate classification with visual explanations that highlight anatomical regions influencing diagnostic decisions."
Disclaimer: Curated by HT Syndication.
