MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641064041 A) filed by Srm University-AP, Amaravati, Andhra Pradesh, on May 21, for 'system and method for predicting heart diseases using hybrid machine learning architecture.'
Inventor(s) include Dr. Lipismita Panigrahi; S Rama Shesha Sai; and T Rohan.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present invention provides a system and method for predicting heart disease using a hybrid dual-architecture framework that integrates a centralized ensemble learning module (100) incorporating multiple machine learning (ML) base models specifically Random Forest (102), Extra Trees (104), Gradient Boosting (106) and Logistic Regression (108) trained on pooled clinical datasets, a meta learner (110) specifically a Ridge Classifier and a cross-validation framework (112) specifically 10-Fold cross-validation for OOF meta-feature generation with a privacy-preserving federated multi-layer perceptron (MLP) learning module (200) trained locally using the four UCI clinical databases (Cleveland, Hungarian, Switzerland, VA), along with an explainable artificial intelligence engine (300) to provide accurate, secure, and interpretable clinical decision support and a comparative loss engine (400) to evaluate performance between centralized ensemble learning module (100) and privacy-preserving federated multi-layer perceptron (MLP) learning module (200)."
Disclaimer: Curated by HT Syndication.
