MUMBAI, India, Feb. 6 -- Intellectual Property India has published a patent application (202641006709 A) filed by Srm University, Guntur, Andhra Pradesh, on Jan. 23, for 'a neuro-symbolic system and a method for quantitative risk scoring from structural mri data.'
Inventor(s) include Greeshma Venkata Siva Sai Bethalam; Rahul Gowtham Poola; Tejaswi Chodavarapu; Raju Galimutti; Ramesh Vaddi; and Siva Sankar Yellampalli.
The application for the patent was published on Feb. 6, under issue no. 06/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure relates to a neuro-symbolic image processing system (100) and method for generating a quantitative risk score from structural magnetic resonance imaging (MRI) data to facilitate interpretable Alzheimer's disease screening. The system (100) comprises a data ingestion and pre-processing module (102) for extracting a 2D hippocampal slice and computing volumetric biomarkers via an automated segmentation pipeline (106). A hybrid neural-symbolic processing engine (104) utilizes a compact CNN (110) for visual features and a neuro-symbolic reasoning unit (130) to compute attention weights over learnable rule vectors (132) representing clinical observations. A fusion and calibration module (140) employs isotonic regression (144) and a logistic regression meta-learner (146) to produce well-calibrated risk scores. A decision support interface (150) compares scores against predetermined thresholds (152) and generates interpretable explanation reports (154). The method ensures clinically trustworthy screening suitable for resource-constrained edge hardware."
Disclaimer: Curated by HT Syndication.
