Related Experiment Video
Updated: Apr 16, 2026
06:53
3D Whole-heart Myocardial Tissue Analysis
Published on: April 12, 2017
9.5K
Segmentation of myocardium from cardiac MR images using a novel dynamic programming based segmentation method.
Xiaohua Qian1, Yuan Lin2, Yue Zhao3
1SJTUCU International Cooperative Research Center, School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiao Tong University, Shanghai 200240, China and Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, China.
Medical Physics
|March 5, 2015
Summary
This study introduces a dynamic programming method for segmenting heart muscle in cardiac MRI scans. The new techniques significantly improve accuracy, making it a valuable tool for diagnosing cardiovascular diseases.
Area of Science:
- Medical Imaging
- Cardiovascular Diseases
- Image Segmentation
Background:
- Myocardium segmentation in cardiac MRI is crucial for diagnosing cardiovascular diseases.
- Low contrast and variations in intensity/shape make myocardium segmentation challenging.
Purpose of the Study:
- Develop a dynamic programming (DP) based method for segmenting myocardium in cardiac MR images.
- Incorporate likelihood and shape information to improve segmentation accuracy.
Main Methods:
- Segment endocardium for initialization, then determine optimal epicardium contour using polar-transformed images and DP.
- Utilize a likelihood image for prior probability estimation, adaptive search range to exclude irrelevant tissues, and a connectivity-constrained DP algorithm for closed contours.
Main Results:
- Qualitative results were clinically acceptable.
- Quantitative evaluation showed mean Dice scores of 0.892 for endocardium and 0.927 for epicardium.
- The three proposed techniques improved epicardium segmentation accuracy (Dice: +0.029 to +0.047; RMSE: -0.21 to -1.31 mm).
Conclusions:
- The three novel techniques enhance the DP method's segmentation capabilities.
- The proposed method achieves acceptable myocardium segmentation in cardiac MR images.
- This method is beneficial for clinical diagnosis of cardiovascular diseases.
