MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641018550 A) filed by R. Kalaivani; Dhivya R; Dr. Gayetri Devi S. V; R. Kiruthiga; Aruloli R; Nandhini J; and Premkumar J, Chennai, Tamil Nadu, on Feb. 18, for 'an energy-efficient federated deep learning model for privacy-preserving healthcare data analytics.'
Inventor(s) include R. Kalaivani; Dhivya R; Dr. Gayetri Devi S. V; R. Kiruthiga; Aruloli R; Nandhini J; and Premkumar J.
The application for the patent was published on Feb. 27, under issue no. 09/2026.
According to the abstract released by the Intellectual Property India: "The present invention proposes an energy-efficient federated deep learning model for privacy-preserving healthcare data analytics, aimed at enabling collaborative intelligence across distributed medical institutions without sharing sensitive patient data. The system employs a federated learning architecture in which healthcare data remain stored and processed locally at hospitals, clinics, and medical devices, while only optimized model updates are exchanged to construct a global deep learning model. To address the high energy consumption and communication overhead associated with conventional federated learning, the proposed approach integrates adaptive energy-aware training strategies, efficient client selection, and compressed parameter transmission. Privacy protection is further strengthened through secure aggregation and privacy-aware update mechanisms that prevent disclosure of individual patient information. By jointly optimizing energy efficiency, data privacy, and analytical accuracy, the proposed invention provides a scalable, secure, and sustainable solution for advanced healthcare data analytics and real-world medical decision support."
Disclaimer: Curated by HT Syndication.
