Related Experiment Video
Updated: Feb 19, 2026
08:52
3D Ultrasound Imaging: Fast and Cost-effective Morphometry of Musculoskeletal Tissue
Published on: November 27, 2017
24.3K
Lossless medical image compression using geometry-adaptive partitioning and least square-based prediction
Xiaoying Song1,2, Qijun Huang3, Sheng Chang2
1Engineering Research Center of Metallurgical Automation and Measurement Technology, Wuhan University of Science and Technology, Wuhan, Hubei, 430081, China.
Medical & Biological Engineering & Computing
|November 7, 2017
Summary
This study introduces an efficient algorithm for lossless medical image compression using irregular segmentation and region-based prediction. The novel method significantly improves compression rates compared to existing standards like JPEG 2000.
Area of Science:
- Medical Imaging
- Computer Science
- Image Processing
Background:
- Lossless compression of medical images is crucial for diagnostic accuracy and data storage.
- Existing compression algorithms face challenges in optimizing compression rates for diverse medical image structures.
Purpose of the Study:
- To develop an efficient algorithm for lossless compression of medical images.
- To enhance compression rates by utilizing irregular segmentation and region-based prediction.
Main Methods:
- A hybrid segmentation method combining geometry-adaptive and quadtree partitioning for adaptive irregular segmentation.
- Design of adaptive least square (LS)-based predictors for each segmented region.
- Exploitation of spatial pixel correlation and local structure similarity.
Main Results:
- The proposed algorithm achieved superior compression performance.
- Average compression improvements of 10.48% over JPEG 2000, 4.86% over CALIC, 3.58% over EDP, and 0.10% over JPEG-LS.
Conclusions:
- The proposed adaptive irregular segmentation and region-based prediction algorithm offers significant improvements in lossless medical image compression.
- This method effectively balances compression efficiency with the preservation of image data integrity.
