Abstract :

Medical Imaging Plays A Crucial Role In Disease Diagnosis And Treatment, Especially With 3D Images Such As Chest Scans That Provide Detailed Anatomical Information. However, These Images Require Significant Storage And Transmission Bandwidth. Traditional Compression Techniques Such As PCA, DWT, And JPEG2000 Often Lead To Quality Degradation During Decompression. To Address These Limitations, This Work Proposes A Federated Learning-based 3D Medical Image Compression System That Combines Optimal Multi-linear Singular Value Decomposition (OMLSVD) With Deep Auto-encoders. In The Proposed Approach, OMLSVD Is Used To Compress 3D Medical Images Efficiently, Reducing Storage Requirements While Preserving Structural Information. The Compressed Images, Along With Original Images, Are Used To Train An Auto-encoder Model That Reconstructs Highquality Images During Decompression. To Ensure Data Privacy, Federated Learning Is Employed, Where Multiple Medical Clients Train The Model Locally And Share Only Model Weights With A Centralized Federated Server Instead Of Raw Data. The System Is Evaluated Using A 3D Chest X-ray Dataset, And Performance Is Measured Using SSIM And PSNR Metrics. Experimental Results Demonstrate That The Proposed Method Achieves Better Compression Efficiency And Higher Reconstruction Quality Compared To Existing Techniques, While Maintaining Data Privacy.

Published:

08-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2027-2034

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite