Related Experiment Video
Updated: Mar 13, 2026
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
43.7K
Efficient liver segmentation in CT images based on graph cuts and bottleneck detection.
Miao Liao1, Yu-Qian Zhao2, Wei Wang3
1School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan 411201, China; School of Information Science and Engineering, Central South University, Changsha 410083, China.
Summary
This study presents an efficient semi-automatic method for segmenting healthy liver regions in CT scans. The technique improves accuracy for liver disease diagnosis and surgical planning.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Image Segmentation
Background:
- Accurate liver segmentation in abdominal CT is crucial for disease diagnosis and surgical planning.
- Challenges include ambiguous edges, tissue adhesion, and variations in liver shape and intensity.
Purpose of the Study:
- To develop an efficient semi-automatic method for segmenting healthy liver regions in CT volumes.
- To address the challenges of complex backgrounds and variations in liver appearance.
Main Methods:
- Combines an intensity model with a principal component analysis (PCA)-based appearance model.
- Integrates location information from neighboring slices into graph cuts for automatic segmentation.
- Employs a boundary refinement method based on bottleneck detection to enhance accuracy.
Main Results:
- The proposed method effectively segments healthy liver regions without extensive training or statistical models.
- Demonstrates capability in handling complex shape and intensity variations.
- Outperforms several existing methods on the XHCSU14 and SLIVER07 databases, validated by MICCAI criteria and Dice similarity coefficient.
Conclusions:
- The developed semi-automatic method offers an efficient and accurate solution for liver segmentation in CT volumes.
- It provides a robust approach for computer-aided liver disease diagnosis and surgical planning.
- The method's performance indicates its potential for clinical application.
