MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641019392 A) filed by Dr. Jenifer Labi Labence; Janani Annur Thiruvengadam; Dhanya G S; J. P. Srividhya; K. S. Kavin; Dr. A. Darcy Gnana Jegha; A. T. R. Krishna Priya; Sherin Paul P; R. Alaguraj; and Dr. S. Abisha, Kanyakumari, Tamil Nadu, on Feb. 19, for 'graph neural network-based classification of brain lesions from mri sequences.'
Inventor(s) include Dr. Jenifer Labi Labence; Janani Annur Thiruvengadam; Dhanya G S; J. P. Srividhya; K. S. Kavin; Dr. A. Darcy Gnana Jegha; A. T. R. Krishna Priya; Sherin Paul P; R. Alaguraj; and Dr. S. Abisha.
The application for the patent was published on Feb. 27, under issue no. 09/2026.
According to the abstract released by the Intellectual Property India: "This invention introduces a GNN-based approach for the classification of brain lesions from MRI sequences, addressing challenges in medical imaging such as complex lesion structures, spatial dependencies, and imaging variations. Unlike traditional deep learning models like CNNs, which struggle to capture global spatial relationships, GNNs represent brain regions as interconnected nodes, preserving anatomical and contextual information. The proposed method integrates multi-sequence MRI data, allowing for a more comprehensive analysis of lesion characteristics. Advanced techniques such as spectral graph convolutions and attention mechanisms enhance feature extraction, improving classification accuracy and robustness. Furthermore, message-passing frameworks refine node interactions, enabling the model to differentiate between various lesion types effectively. This innovation significantly enhances the speed and precision of automated diagnosis, assisting radiologists in making informed clinical decisions. By incorporating AI-driven solutions into neuroimaging workflows, the invention contributes to improved patient care, early diagnosis, and optimized treatment planning."
Disclaimer: Curated by HT Syndication.
