MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641075657 A) filed by Sr University on June 18, 2026, for A Multi-Stage Deep Learning System For Automated Brain Tumor Detection And Classification.
Inventors include Nattala Subbarayudu; Sreedhar Kollem; and T. Venkatakrish Namoorthy.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: A MULTI-STAGE DEEP LEARNING SYSTEM FOR AUTOMATED BRAIN TUMOR DETECTION AND CLASSIFICATION The invention discloses a multi stage deep learning system for automated brain tumor detection and classification from MRI images. The system integrates detection, segmentation, subtype classification, 3D modeling, and explainability into a unified pipeline. Tumor presence is identified using convolutional neural networks, while segmentation is performed through a hybrid approach combining CNN feature extraction with fuzzy C means clustering. Classified tumors are reconstructed into volumetric 3D models, with quantitative features such as volume, shape, and anatomical location computed for clinical decision support. Explainable AI modules including Grad CAM, LIME, and SHAP generate heatmaps and feature importance maps, providing transparency and interpretability of predictions. The system generalizes across multiple MRI modalities and institutional protocols, offering improved accuracy, reduced manual variability, and actionable outputs for surgical planning and prognosis. This invention represents a robust, automated, and clinically interpretable solution for brain tumor analysis.
Disclaimer: Curated by HT Syndication.
