Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195304,
        author = {A Vinayakasai and A Devi Krishna and M Pranay and M Sai Yeshwanth and Dr. S Shiva Prasad},
        title = {Federated Learning- Based 3D Medical Image Compression},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {7866-7875},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195304},
        abstract = {The rapid growth of three-dimensional medical imaging modalities such as Magnetic Resonance Imaging and Computed Tomography has significantly increased the volume of healthcare data generated by modern medical systems. Conventional image compression techniques and centralized deep learning approaches require direct access to raw medical data, leading to challenges related to patient privacy, regulatory compliance, and secure data sharing. These limitations restrict large-scale collaboration across healthcare institutions and increase the risk of data exposure during storage and transmission. To address these challenges, the proposed approach integrates federated learning with deep neural network–based compression to enable efficient and privacy-preserving 3D medical image compression. In this framework, multiple medical institutions collaboratively train local 3D convolutional autoencoder models without exchanging sensitive patient data. Only encrypted model updates are shared and aggregated using federated averaging to construct a global compression model. Performance evaluation using compression ratio, Peak Signal-to-Noise Ratio, and Structural Similarity Index Measure demonstrates effective compression while preserving diagnostic quality. The results indicate improved data privacy, and enhanced scalability, making the proposed framework suitable for secure and distributed healthcare environments},
        keywords = {Distributed Deep Learning, Medical Imaging, Convolutional Autoencoders,3D Convolutional Neural Networks, Data Privacy and Security, Decentralized Model Training},
        month = {March},
        }