MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641095961 A) filed by St Joseph Engineering College on August 07, 2026, for A System And Method For Segmenting And Analyzing Medical Images Using Multi-Modal Fusion.
Inventors include Supriya Salian; Vamshith S Bangera; Rashmi S D; and Sneha.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: The present invention relates to a system (100) and a method (300) for segmenting and analyzing medical images. The method (300) comprises acquiring, by a processor (201), visual data of a medical image comprising a plurality of frames captured by a visual data capturing unit (209), extracting multi-modal feature representations from each frame using a first encoder and a second encoder of a tri-modal fusion architecture, and aligning the feature representations into a shared embedding space through learnable projection heads using a contrastive learning framework comprising a bidirectional loss function scaled by a temperature parameter. The method (300) further comprises fusing the aligned feature representations to produce a fused feature map, generating a binary segmentation mask, estimating a size of the identified cancerous tissue region in clinical units, classifying a cancer stage, and generating a structured medical report for transmission to a remote device (103) via a network (101). FIG.2
Disclaimer: Curated by HT Syndication.
