MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202641123341 A) filed by S. Hrushikesava Raju; Dr. Suneetha Bulla; and Koneru Lakshmaiah Education Foundation on October 14, 2026, for A Deep Learning Based System And Method For Automated Detection And Classification Of Brain Tumors From Magnetic Resonance Imaging (mri) Using Resnet50 And Transfer Learning.
Inventors include Muvvala Venkata Bala Srunith; Kollisetty Sai Krishna; Pallem Chethana; Talapaneni Harsha Vardhan; and Dr. Suneetha Bulla.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: A computer-implemented system and method are disclosed for the automated detection and classification of brain tumors from Magnetic Resonance Imaging (MRI) images. An image preprocessing module performs skull stripping, multi-scanner intensity normalization, resizing and data augmentation on an input MRI image. A feature extraction module comprising a pre-trained ResNet50 convolutional neural network, adapted through transfer learning, extracts hierarchical features from the preprocessed image. A classification module comprising a dense classification head classifies the extracted features into a 'Tumor' or 'Non-Tumor' category with an associated confidence score. An evaluation and benchmarking module computes accuracy, precision, recall, F1-score, confusion matrix and AUC (ROC), and presents these together with the classification output, via a reporting interface, to a clinician, additionally benchmarking the computed metrics against comparative prior-art ResNet50-based approaches reporting accuracies of up to 99.88%. The disclosed system provides an efficient, reliable and deployable computer-aided tool for brain tumor detection from MRI images, intended to assist, and not replace clinical diagnosis.
Disclaimer: Curated by HT Syndication.
