Related Experiment Video
Updated: Jul 31, 2025
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
Published on: December 15, 2023
Graph matching and deep neural networks based whole heart and great vessel segmentation in congenital heart disease
Zeyang Yao1,2, Wen Xie1,2, Jiawei Zhang3
1School of Medicine, South China University of Technology, Guangzhou, 510006, China.
Insights
This study introduces a novel deep learning and graph algorithm framework for segmenting complex congenital heart disease (CHD) in CT images. The method significantly improves segmentation accuracy, paving the way for clinical applications.
Area of Science:
- Medical imaging
- Computer-aided diagnosis
- Cardiovascular research
Background:
- Congenital heart disease (CHD) is a major cause of infant mortality.
- Accurate segmentation of the heart and great vessels in CHD is challenging due to anatomical variations.
- Existing segmentation methods struggle with the complexity of CHD structures.
Purpose of the Study:
- To develop an automated framework for segmenting the whole heart and great vessels in complex CHD using CT images.
- To combine deep learning for regular structures and graph algorithms for variations.
- To improve the accuracy and clinical applicability of CHD segmentation.
Main Methods:
- A hybrid framework utilizing deep learning for chamber and myocardium segmentation.
- Graph matching algorithms applied to vessel connection information for categorization.
- Validation on 68 3D CT datasets covering 14 types of CHD.
Main Results:
- The proposed framework achieved an average 12% increase in Dice score compared to state-of-the-art methods for normal anatomy.
- Clinical evaluation by cardiovascular imaging specialists demonstrated good performance against the Van Praagh classification system.
- The method shows promise for accurate segmentation of complex CHD structures.
Conclusions:
- The combined deep learning and graph algorithm approach effectively addresses challenges in CHD segmentation.
- This framework offers a significant improvement over existing methods for whole heart and great vessel segmentation in CHD.
- The study paves the way for future clinical integration of this automated segmentation tool.
Abstract:
Congenital heart disease (CHD) is one of the leading causes of mortality among birth defects, and due to significant variations in the whole heart and great vessel, automatic CHD segmentation using CT images has been always under-researched. Even though some segmentation algorithms have been developed in the literature, none perform very well under the complex structure of CHD. To deal with the challenges, we take advantage of deep learning in processing regular structures and graph algorithms in dealing with large variations and propose a framework combining both the whole heart and great vessel segmentation in complex CHD. We benefit from deep learning in segmenting the four chambers and myocardium based on the blood pool, and then we extract the connection information and apply graph matching to determine the categories of all the vessels. Experimental results on 68 3D CT images covering 14 types of CHD illustrate our framework can increase the Dice score by 12% on average compared with the state-of-the-art whole heart and great vessel segmentation method in normal anatomy. We further introduce two cardiovascular imaging specialists to evaluate our results in the standard of the Van Praagh classification system, and achieves well performance in clinical evaluation. All these results may pave the way for the clinical use of our method in the incoming future.
