MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641103808 A) filed by Chennai Institute Of Technology on August 28, 2026, for Health Monitoring System For Blood Flow Abnormality Detection Using Hybrid Machine Learning Models.
Inventors include Dr. P. Suresh Kumar; Dr. S. Meenakshi; A. Kumaresan; and S. Sadhana.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The titled invention discloses an health monitoring system for blood flow abnormality 5 detection using hybrid machine learning models. The system processes a retinal fundus image (1) through an image preprocessing unit (2) using Contrast Limited Adaptive Histogram Equalization (CLAHE) and Gaussian filtering for enhancing vascular structures and reducing illumination variations. The preprocessed image is supplied to an Attention U-Net segmentation unit (3) having attention gates for 1.0 focusing on vessel-related regions and suppressing irrelevant information. A residual feature extraction unit (4) extracts contextual and fine-grained vessel features while mitigating the vanishing gradient problem. The resulting segmented vessel map (5) represents the detected retinal blood vessel structures. A performance evaluation unit (6) evaluates the segmentation using statistical metrics, ROC curve, diagnostic curve, 15 and Dice coefficient. The method is trained using the Adam optimizer with binary cross entropy loss and evaluated using the DRIVE database, providing an automated approach for retinal vessel segmentation and supporting retinal disease detection and analysis.
Disclaimer: Curated by HT Syndication.
