MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641076341 A) filed by Sr University on June 19, 2026, for A Deep Learning-Based Stroke Lesion Localization And Segmentation System.
Inventors include Burgula Sowmya; Dr. Johnson Kolluri; Dr. P. Shailaja; and Dr. K. Praveen Kumar.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: ABSTRACT Disclosed herein is a deep learning-based stroke lesion localization and segmentation system (100), the system (100) comprises an image acquisition module (102) configured to receive MRI and/or CT medical imaging data. A preprocessing module (104) performs image normalization, noise reduction, skull stripping, image enhancement, registration, and data augmentation. A feature extraction module (106) automatically extracts spatial, structural, and pathological features, while a lesion localization module (108) identifies stroke-affected brain regions. A deep learning segmentation module (110) utilizing an encoder-decoder neural network architecture. A classification module (112) distinguishes ischemic stroke lesions, hemorrhagic stroke lesions, and other intracranial abnormalities. A model training and optimization module (114) trains the deep learning architecture. A lesion quantification module (116) determines lesion size, volume, location, and extent. A performance evaluation module (118) assesses accuracy. A clinical decision support module (120) presents lesion localization, segmentation results, quantification information, and diagnostic recommendations.
Disclaimer: Curated by HT Syndication.
