Title: Secure and privacy-preserving medical image retrieval model in cloud-edge computing

Authors: Sirisha Potluri; R. Balamurali; B. Ravikrishna

Addresses: Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Bowrampet, Hyderabad, Telangana, India ' Department of Computer Science and Engineering, Faculty of Science and Technology, The ICFAI Foundation for Higher Education, Hyderabad, Telangana, India ' Department of Artificial Intelligence and Data Science, Vignan Institute of Technology and Science, Hyderabad, Telangana, India

Abstract: Content-based image retrieval is one of the significant image retrieval techniques. There is a challenging task for content-based image retrieval due to the risk of privacy concerns, leakage of sensitive data and encryption of the images before keeping them in the cloud. In this work, we propose a secure and privacy-preserving medical image retrieval model to support multiple image sources or owners and their sharing concerns. We securely encrypt the image features with a multi-party computation and sharing model with their keys. This allows us to achieve efficient image retrieval over several images that are collected from multiple sources with guaranteed image privacy. Using this technique, we can also get similarity measurements of the images to avoid keeping image similarity information in the cloud. Results demonstrate that our proposed model achieved improved image retrieval accuracy while preserving the security and privacy of the image.

Keywords: content-based image retrieval; healthcare; computer vision; cloud computing; edge computing; encryption.

DOI: 10.1504/IJGUC.2025.148547

International Journal of Grid and Utility Computing, 2025 Vol.16 No.5/6, pp.472 - 480

Received: 24 Jun 2024
Accepted: 01 Aug 2024
Published online: 11 Sep 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article