MedSR-QANN: A Next-Generation Framework for Intelligent Medical Image Compression

Abstract

The growing use of digital healthcare systems has led to a significant increase in the generation and transmission of high-resolution medical images, creating challenges in storage management and network bandwidth utilization. Efficient image compression techniques are therefore essential to reduce data size while preserving clinically relevant information. This study presents a hybrid medical image compression framework that combines Quantum-Enhanced Artificial Neural Networks (QANN) with a Super-Resolution (SR) enhancement module to improve reconstruction quality after compression. The QANN model is employed to learn compact feature representations and perform efficient image compression and reconstruction. To address the loss of fine anatomical details that may occur at higher compression levels, a Super-Resolution mechanism is integrated into the post-reconstruction stage. The SR module enhances image sharpness, restores subtle structural features, and suppresses reconstruction artifacts, resulting in improved visual and diagnostic quality. The proposed system is implemented through a user-friendly Flask-based web application that enables image upload, compression, reconstruction, and enhancement in a seamless workflow. Experimental evaluation demonstrates that the integration of Super-Resolution significantly improves reconstructed image quality while maintaining effective compression performance. The proposed framework offers a practical solution for medical image storage and transmission, making it particularly suitable for telemedicine, remote healthcare services, and other bandwidth-limited clinical environments. Index terms - — Quantum-Enhanced Artificial Neural Network (QANN), Medical Image Compression, Super Resolution, Quantum Feature Extraction, Hybrid Quantum–Classical Model, Image Reconstruction, Telemedicine.