MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202641082067 A) filed by Chaitanya Bharathi Institute Of Technologya on July 03, 2026, for Decentralized Federated Learning With Blockchain For Patient Data Privacy.
Inventors include Tarun Nellikuduru; Yaswanth Simha Jinkathoti; Navya Sri Koppula; Dr. Ramu Kuchipudi; Dr. M Venu Gopalachari; and Dr. Sagar Gujjunoori.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: DECENTRALIZED FEDERATED LEARNING WITH BLOCKCHAIN FOR PATIENT DATA PRIVACY The present invention discloses a decentralized federated learning framework integrated with blockchain and InterPlanetary File System (IPFS) for ensuring secure and privacy-preserving collaborative model training, particularly for sensitive patient data. The system enables multiple healthcare institutions to collaboratively train machine learning models without sharing raw data, thereby maintaining data confidentiality and regulatory compliance. Blockchain technology is employed to provide a tamper-proof and transparent ledger for managing participant identities, model update transactions, and validation processes through smart contracts. IPFS is utilized for decentralized storage of encrypted model updates, reducing storage overhead and enhancing system scalability. The invention eliminates the need for a central aggregator by enabling decentralized validation and aggregation of model parameters. It incorporates secure authentication, encryption mechanisms, and automated verification to prevent unauthorized access and malicious contributions. The global model is iteratively updated and redistributed among participants, ensuring continuous learning and improved model accuracy. The proposed framework enhances trust, fault tolerance, and efficiency while addressing challenges related to data privacy, security, and scalability in distributed healthcare environments. FIG.1.
Disclaimer: Curated by HT Syndication.
