Related Experiment Video
Updated: Jun 26, 2026
08:41
Novel In Vivo Micro-Computed Tomography Imaging Techniques for Assessing the Progression of Non-Alcoholic Fatty Liver Disease
Published on: March 24, 2023
An Improved Level Set for Liver Segmentation and Perfusion Analysis in MRIs.
Gang Chen1, Lixu Gu, Lijun Qian
1Image Guided Surgery and Therapy Laboratory, Med-X Research Institute and Department of Computer Science, Shanghai Jiao Tong University, China.
Summary
Accurate liver segmentation from MRIs is vital for liver perfusion analysis. A novel multiple-step level set method (LSM) improves segmentation accuracy, overcoming issues with artifacts and low gradients for better perfusion curve analysis.
Area of Science:
- Medical Imaging
- Image Processing
- Radiology
Background:
- Accurate liver segmentation is essential for automated liver perfusion analysis.
- Existing methods like level set methods (LSMs) struggle with artifacts and low gradients, leading to unsatisfactory segmentation.
- Liver perfusion analysis provides critical information about liver blood supply.
Purpose of the Study:
- To develop an improved liver segmentation method for MRI.
- To overcome limitations of current segmentation techniques, specifically leakage and over-segmentation.
- To enable more accurate liver perfusion analysis by refining the segmentation process.
Main Methods:
- Proposed a multiple-initialization, multiple-step level set method (LSM).
- Employed fast marching methods and LSMs for initial contour evolution.
- Integrated a convex hull algorithm for rough contour generation, followed by global level set smoothing for precise boundary determination.
Main Results:
- The proposed multiple-step LSM achieved superior liver segmentation compared to existing methods.
- Experimental results on 12 abdominal MRI series demonstrated enhanced segmentation accuracy.
- Refined liver perfusion curves, free from respiratory artifacts, were obtained using a modified chamfer matching algorithm.
Conclusions:
- The developed multiple-initialization, multiple-step LSM effectively addresses segmentation challenges in liver MRI.
- The improved segmentation facilitates more accurate and reliable liver perfusion analysis.
- This advancement holds potential for improved clinical assessment of liver conditions.
