MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641115452 A) filed by Sri Eshwar College Of Engineering on September 26, 2026, for Explainable Ai Framework For Early Detection And Risk Stratification Of Chronic Diseases Using Heterogeneous Health Data.
Inventors include R Arun; D Monika Sree; S Sabarikarthika; and D K Dhevavarshana.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: The graphical abstract illustrates the comprehensive architectural pipeline of the proposed Explainable AI framework, delineating the systematic progression from multi-modal patient data ingestion to actionable clinical decision support. The workflow initiates at the heterogeneous data ingestion layer, where twenty clinical parameters spanning socio-demographics, physiological vitals, laboratory biomarkers, and behavioral lifestyle indices are gathered and routed into a unified data preprocessing pipeline that performs missing value imputation, biological boundary clipping, and z-score feature standardization. The normalized feature tensor is then processed by the multi-model machine learning inference engine, which benchmarks candidate algorithms including regularized Logistic Regression, Support Vector Machines, Random Forests, and XGBoost to predict calibrated multi-morbidity risk probabilities. These computed probabilities bifurcate simultaneously into two complementary analytical engines: a three-tier risk stratification module that categorizes patient risk into low, moderate, and high tiers, and an Explainable AI module powered by cooperative game-theoretic Shapley Additive exPlanations (SHAP) that decomposes the prediction into directional positive (risk-elevating) and negative (protective) feature contributions. These local attributions are directly synthesized by an evidence-based clinical recommendation rule engine that maps top risk drivers to targeted nutritional, physical activity, surveillance, and specialist referral guidelines conforming to international clinical standards. Finally, the synthesized outputs are rendered onto an interactive healthcare professional dashboard featuring semi-doughnut risk gauges, dynamic SHAP waterfall plots, and a longitudinal audit registry that tracks patient risk trajectories over successive clinical encounters.
Disclaimer: Curated by HT Syndication.
