MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641075749 A) filed by Btp Madhav; and Koneru Lakshmaiah Education Foundation on June 18, 2026, for An Explainable Deep Learning Framework For Brain Tumor Classification Using Mri Images.
Inventors include B. Rakesh Babu; V Rajesh; and B T P Madhav.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: Brain tumor classification from MRI images is essential for accurate diagnosis and treatment planning. This work proposes an Explainable Hybrid Feature Learning Framework that integrates image preprocessing, tumor segmentation, multi-feature extraction, adaptive feature fusion, and automated classification for reliable brain tumor diagnosis. The framework combines complementary feature representations from segmented tumor regions and employs an explainability mechanism to provide interpretable classification outcomes, thereby enhancing clinical trust and decision-making. Experimental evaluation on benchmark MRI datasets achieved a Dice Score of 98.1%, IoU of 96.4%, Accuracy of 99.2%, Precision of 99.0%, Recall of 99.1%, Specificity of 99.4%, F1-Score of 99.0%, and ROC-AUC of 0.998. The obtained results demonstrate that the proposed framework provides highly accurate, robust, and explainable brain tumor classification, making it suitable for intelligent computer-aided diagnostic applications.
Disclaimer: Curated by HT Syndication.
