MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202621065659 A) filed by Nandini Sharma; and Dr Pritaj Yadav on May 25, 2026, for A Secure And Efficient Framework For Enhancing Data And Model Security In Ai-Driven Healthcare.
Inventors include Nandini Sharma; and Dr Pritaj Yadav.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: A secure and efficient framework for AI-driven healthcare systems is disclosed, integrating advanced encryption, digital watermarking, and machine learning to address critical challenges in data confidentiality and model integrity. The framework ensures that sensitive healthcare data, including Electronic Health Records (EHRs), diagnostic images, and sensor-generated patient data, remain encrypted during storage, transmission, and processing using AES-256 encryption. A dual-layer architecture is employed, combining data security with model protection through homomorphic encryption and watermarking techniques. Digital watermarking embeds traceable identifiers within model parameters to prevent unauthorized replication and tampering without affecting predictive performance. Experimental evaluation demonstrates minimal computational overhead, efficient encryption–decryption performance, and robust resistance against attacks, achieving a test accuracy of 76.40%. The proposed framework ensures scalability, regulatory compliance, and suitability for real-time, latency-sensitive healthcare environments, thereby enabling secure, reliable, and trustworthy deployment of artificial intelligence in modern healthcare systems.
Disclaimer: Curated by HT Syndication.
