MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202621081989 A) filed by Symbiosis International Deemed University on July 02, 2026, for An Ai-Based System For Brain Tumor Detection And Classification Using Mri Images.
Inventors include Madhur Wanjari; Vansh Mahant; Ojas Jibhkate; and Dr. Priya Dasarwar.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: ABSTRACT AN AI-BASED SYSTEM FOR BRAIN TUMOR DETECTION AND CLASSIFICATION USING MRI IMAGES The present invention discloses an enhanced artificial intelligence based system (100) for brain tumor detection and classification using deep learning on MRI images. The system (100) comprises an image acquisition module (110) for receiving MRI brain scans; a preprocessing module (120) for normalization, resizing, noise removal and data augmentation; a feature extraction engine (130) for extracting shape, statistical and texture features; a deep learning classification engine (140) employing Convolutional Neural Networks and transfer learning architectures including VGG16 and ResNet50 for binary tumor detection and multi-class classification into glioma, meningioma and pituitary tumor types; a performance evaluation module (150) for computing accuracy, precision, recall, F1-score and confusion matrix metrics; and an explainable AI visualization module (160) incorporating Gradient-weighted Class Activation Mapping (Grad-CAM) for generating heatmap overlays that highlight tumor-indicative regions. The system achieves classification accuracy of up to 96 percent with model prediction in less than one second. [
Disclaimer: Curated by HT Syndication.
