Related Experiment Video
Updated: Oct 6, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
Vessel segmentation for X-ray coronary angiography using ensemble methods with deep learning and filter-based
Zijun Gao1, Lu Wang2, Reza Soroushmehr2,3,4
1Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, USA. zijung@umich.edu.
Insights
This study introduces a novel ensemble framework for segmenting coronary arteries in X-ray coronary angiography (XCA) images, improving computer-aided diagnosis of coronary artery disease (CAD). The method shows superior performance over deep learning models, enhancing patient care.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Healthcare
- Cardiovascular Disease Diagnostics
Background:
- Automated segmentation of coronary arteries is vital for computer-aided diagnosis and treatment planning of coronary artery disease (CAD).
- Accurate delineation in X-ray coronary angiography (XCA) is challenging due to low signal-to-noise ratio and background interference.
Purpose of the Study:
- To propose a novel ensemble framework for coronary artery segmentation in XCA images.
- To enhance the accuracy and consistency of automated segmentation compared to existing deep learning methods.
Main Methods:
- Developed an ensemble framework combining deep learning and filter-based features using Gradient Boosting Decision Tree (GBDT) and deep forest classifiers.
- Constructed 37-dimensional feature vectors from multi-scale filtering responses and deep neural network feature maps.
- Trained and tested the models on 130 XCA images, evaluating performance using precision, sensitivity, specificity, F1 score, AUROC, and IoU.
Main Results:
- The best GBDT model achieved an F1 score of 0.874 and AUROC of 0.947.
- The best deep forest model achieved an F1 score of 0.867 and AUROC of 0.95.
- Both ensemble models demonstrated superior or comparable performance with lower standard deviations than common deep neural networks.
Conclusions:
- The proposed feature-based ensemble method outperforms standard deep convolutional neural networks in coronary artery segmentation.
- This method provides more consistent results, facilitating stenosis assessment and improving care for CAD patients.
Background:
Automated segmentation of coronary arteries is a crucial step for computer-aided coronary artery disease (CAD) diagnosis and treatment planning. Correct delineation of the coronary artery is challenging in X-ray coronary angiography (XCA) due to the low signal-to-noise ratio and confounding background structures.
Methods:
A novel ensemble framework for coronary artery segmentation in XCA images is proposed, which utilizes deep learning and filter-based features to construct models using the gradient boosting decision tree (GBDT) and deep forest classifiers. The proposed method was trained and tested on 130 XCA images. For each pixel of interest in the XCA images, a 37-dimensional feature vector was constructed based on (1) the statistics of multi-scale filtering responses in the morphological, spatial, and frequency domains; and (2) the feature maps obtained from trained deep neural networks. The performance of these models was compared with those of common deep neural networks on metrics including precision, sensitivity, specificity, F1 score, AUROC (the area under the receiver operating characteristic curve), and IoU (intersection over union).
Results:
With hybrid under-sampling methods, the best performing GBDT model achieved a mean F1 score of 0.874, AUROC of 0.947, sensitivity of 0.902, and specificity of 0.992; while the best performing deep forest model obtained a mean F1 score of 0.867, AUROC of 0.95, sensitivity of 0.867, and specificity of 0.993. Compared with the evaluated deep neural networks, both models had better or comparable performance for all evaluated metrics with lower standard deviations over the test images.
Conclusions:
The proposed feature-based ensemble method outperformed common deep convolutional neural networks in most performance metrics while yielding more consistent results. Such a method can be used to facilitate the assessment of stenosis and improve the quality of care in patients with CAD.
Related Concept Videos
01:20Imaging Studies for Cardiovascular System III: X-Ray
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...
01:19Imaging Studies VII: Vascular Imaging
01:28Imaging Studies for Cardiovascular System V: CT