MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641104730 A) filed by Vardhaman College Of Engineering on August 31, 2026, for Federated Meta-Learning For Personalized Healthcare Predictions Across Distributed Systems.
Inventors include Mr. K. Anvesh; Dr. L. Sunitha; Dr. K. Nikhila; Mr. Dastagiri Reddy Pichala; Dr. Srinivasulu Gogula; and Mr S Satheesh Kumar.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: Federated Meta-Learning for Personalized Healthcare Predictions Across Distributed Systems is the proposed invention. The proposed architecture combines federated learning, adaptive meta-learning, attention-based personalisation, secure aggregation, differential privacy and continual learning to enable collaborative model training without sharing sensitive medical data. Each participating healthcare institution trains local models on private clinical datasets and shares only encrypted model parameters with a federated server, where an adaptive aggregation yields an optimised global model, which can quickly adapt to different patient populations and changing clinical environments. The framework well addresses the challenges of data heterogeneity, small labelled data and privacy concerns faced in healthcare applications. The invention combines personalised learning with secure distributed intelligence to enhance disease prediction, treatment recommendation, patient risk assessment, chronic disease management, and clinical decision support with regulatory compliance, scalability, robustness, and data confidentiality across geographically distributed healthcare networks.
Disclaimer: Curated by HT Syndication.
