MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202621066961 A) filed by Ms. Swatee Subhash Nikam; and Dr. Nilima Kulkarni on May 28, 2026, for A System And Method For Privacy-Preserving Medical Image Classification Using Ckks-Based Encrypted Data Handling And Lightweight Convolutional Neural Network.
Inventors include Ms. Swatee Subhash Nikam; and Dr. Nilima Kulkarni.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: The present invention relates to a computer-implemented system and method for privacy-preserving medical image classification using CKKS-based encrypted data handling and a lightweight convolutional neural network. The system receives a medical image from an authorized healthcare source and preprocesses the image into a compact normalized numerical representation. The preprocessed representation is encoded and encrypted using a CKKS-based encryption context to generate a ciphertext representation before storage or transmission. The encrypted representation is stored or transmitted in protected form so that the raw medical image or its directly readable numerical form is not exposed to untrusted storage systems, communication channels, or cloud-assisted environments. For diagnostic classification, the protected image representation is recovered only within an authorized trusted inference boundary. The recovered image data is then supplied to a lightweight convolutional neural network configured to classify the medical image into one or more diagnostic categories. The system generates and discloses only the final diagnostic class output, and optionally a confidence score, to an authorized user or medical operator. The invention integrates preprocessing, CKKS-based encrypted data handling, secure storage and transmission, trusted-zone recovery, lightweight CNN classification, and output-only disclosure into a single privacy-preserving medical image classification workflow suitable for hospitals, diagnostic centers, tele-radiology systems, and secure healthcare artificial intelligence platforms.
Disclaimer: Curated by HT Syndication.
