MUMBAI, India, Nov. 21 -- Intellectual Property India has published a patent application (202541085291 A) filed by Manipal Academy Of Higher Education, Manipal, Karnataka, on Sept. 9, for 'a system and method for multimodal deep learning diagnosis of diabetic neuropathy.'
Inventor(s) include Mehewish Musheer Sheikh; Mamatha Balachandra; Narendra V G; and G Arun Maiya.
The application for the patent was published on Nov. 21, under issue no. 47/2025.
According to the abstract released by the Intellectual Property India: "The present invention relates to an automated diagnostic system for classifying the severity of diabetic peripheral neuropathy by integrating plantar pressure imaging with structured clinical data using an early fusion deep learning architecture. The system comprises an image branch for extracting biomechanical features, a tabular branch for processing clinical variables, and a fusion module that combines both modalities at the feature level to enhance diagnostic accuracy. A classification module categorizes neuropathy severity into mild, moderate, and severe, while an explainability module generates Grad-CAM heatmaps and clinical feature importance rankings to ensure transparency. The invention addresses the limitations of unimodal and late fusion methods by providing improved accuracy, robust generalization, and confidence-aware outputs. A user-friendly interface delivers interpretable results, enabling adoption in clinical workflows and supporting preventive screening, remote monitoring, and broader biomedical applications."
Disclaimer: Curated by HT Syndication.
