MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202611073880 A) filed by Swami Rama Himalayan University on June 15, 2026, for System For Glioma Detection And Segmentation From Multi-Format Mri And Deep Learning.
Inventors include Babita Negi; Manisha Pundir; Nisha Pundir; Ritu Rani; Nisha; Dr. Anupama Mishra; and Dr. Bineet Kumar Joshi.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: The invention relates to medical image analysis and, more particularly, to a deep learning-based system and method for multi-format glioma detection and tumor segmentation from brain magnetic resonance imaging data. A data uploading block receives studies in multiple formats, and a preprocessing block standardizes the studies through format harmonization, volumetric loading, ordered slice extraction, resizing, grayscale conversion, noise reduction, and intensity normalization. A deep learning model block, including a convolutional neural network, generates slice-wise glioma probability scores that are aggregated into a study-level glioma detection output. Responsive thereto, a segmentation block, preferably an encoder-decoder network with skip connections, generates a pixel-wise or region-wise tumor mask, including labelled tumor regions. An output generation block produces a confidence-linked result including the study-level glioma detection output, confidence value, tumor mask, and overlay image, with optional quantitative tumor area estimation and multi-modal sequence handling.
Disclaimer: Curated by HT Syndication.
