Region of Interest based Medical Image Compression

2025-01-06

AI Summary

To address the conflict between massive medical image transmission and diagnostic-level fidelity preservation of critical lesions in telemedicine, this paper proposes a ROI-aware compression method integrating UNet with HEVC. First, a UNet model trained on BraTS 2020 precisely segments tumor regions; subsequently, the segmentation mask drives HEVC’s ROI coding via adaptive QP mapping, enabling lossless or near-lossless reconstruction of lesion regions (PSNR > 42 dB) while applying high compression to background areas. This work pioneers end-to-end co-optimization of semantic segmentation and HEVC ROI coding, overcoming the traditional global rate-distortion trade-off limitation. Experimental results demonstrate a 37% bit-rate reduction over standard HEVC, significantly lowering storage and bandwidth requirements—thereby satisfying the dual demands of diagnostic-grade fidelity and real-time transmission for remote clinical consultation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Security and Privacy: Data transparency and provenance

Abstract

The vast volume of medical image data necessitates efficient compression techniques to support remote healthcare services. This paper explores Region of Interest (ROI) coding to address the balance between compression rate and image quality. By leveraging UNET segmentation on the Brats 2020 dataset, we accurately identify tumor regions, which are critical for diagnosis. These regions are then subjected to High Efficiency Video Coding (HEVC) for compression, enhancing compression rates while preserving essential diagnostic information. This approach ensures that critical image regions maintain their quality, while non-essential areas are compressed more. Our method optimizes storage space and transmission bandwidth, meeting the demands of telemedicine and large-scale medical imaging. Through this technique, we provide a robust solution that maintains the integrity of vital data and improves the efficiency of medical image handling.

Problem

• Medical Image Compression
• Information Integrity
• Telemedicine

Innovation

• ROI Recognition
• HEVC Encoding
• UNET Algorithm