MUMBAI, India, March 13 -- Intellectual Property India has published a patent application (202641025359 A) filed by Bala Bhoomeshwar, Ghatkesar, Telangana, on March 3, for 'explainable multi stage non-invasive anemia detection system using adaptive hybrid local global vision architecture.'
Inventor(s) include Jogu Kalyani; Bala Bhoomeshwar; Muntha Raju; Bakaram Keerthi; Charlapalli Swathi; and Konda Akhil.
The application for the patent was published on March 13, under issue no. 11/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses an explainable multi stage non-invasive anemia detection system based on digital nail image analysis. The system comprises an image acquisition module, a preprocessing module for nail region localization and illumination normalization, and an adaptive hybrid local-global vision architecture. The hybrid architecture integrates a local feature extraction network for spatial and texture analysis with a global contextual modeling network for capturing long-range dependencies within the nail region. A dynamic feature fusion mechanism combines local and global representations to generate a hybrid feature vector. The fused representation is processed by a multi-stage classification module configured to predict anemia severity levels including normal, mild, moderate, and severe. An explainability module generates visual interpretability maps highlighting diagnostically significant regions influencing the prediction. The invention provides a non-invasive, scalable, and deployable solution for anemia screening without requiring blood sample analysis."
Disclaimer: Curated by HT Syndication.
