Related Experiment Video
Updated: Mar 11, 2026
12:09
Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations
Published on: January 8, 2013
14.1K
Dynamic updating atlas for heart segmentation with a nonlinear field-based model.
Ken Cai1, Rongqian Yang2, Hongwei Yue3
1School of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou, 510225, China.
Summary
This study introduces a novel dynamic updating atlas algorithm for accurate cardiac computed tomography (CT) segmentation. The method enhances heart and lung function assessment using nonlinear deformation fields for precise 4D cardiac segmentation.
Area of Science:
- Medical Imaging
- Cardiovascular Imaging
- Image Analysis
Background:
- Cardiac computed tomography (CT) segmentation is vital for assessing heart and lung function.
- Atlas-based segmentation quality depends heavily on reference image selection.
- Optimal reference images closely match target images for accurate segmentation.
Purpose of the Study:
- To develop a dynamic updating atlas algorithm for improved cardiac CT segmentation.
- To leverage nonlinear deformation fields for precise 4D cardiac segmentation.
- To enhance the accuracy of atlas-based segmentation methods.
Main Methods:
- Proposed an atlas dynamic update algorithm utilizing nonlinear deformation fields.
- Extracted features from double-source CT (DSCT) slices to build an average model.
- Dynamically updated reference atlas images during registration for 4D cardiac segmentation.
Main Results:
- Validated the segmentation framework on 14 4D cardiac CT sequences.
- Achieved acceptable segmentation accuracy ranging from 1.0 to 2.8 mm.
- Demonstrated the effectiveness of the nonlinear field-based model in 4D segmentation.
Conclusions:
- The proposed method offers an effective and accurate approach for whole heart segmentation.
- Combines nonlinear field-based modeling with dynamic atlas updating strategies.
- Success relies on effective atlas prior knowledge and similarity exploration in DSCT sequences.
