Related Experiment Video
Updated: Jul 23, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries
Şerife Kaba1, Huseyin Haci2, Ali Isin3
1Department of Biomedical Engineering, Near East University, TRNC Mersin 10, Nicosia 99138, Turkey.
Insights
This study introduces deep learning for coronary artery disease (CAD) detection. U-Net excels at artery segmentation, while DenseNet201 accurately classifies stenosis, aiding cardiologists and reducing errors.
Area of Science:
- Cardiology
- Medical Imaging Analysis
- Artificial Intelligence in Healthcare
Background:
- Coronary artery disease (CAD) is a leading global cause of mortality.
- Accurate detection of coronary artery stenosis is critical for effective treatment.
- Current manual interpretation of coronary angiography is prone to high workloads, delays, and human error.
Purpose of the Study:
- To develop and evaluate deep learning models for automated segmentation and classification of coronary arteries.
- To compare the performance of U-Net, ResUNet-a, and UNet++ for coronary artery segmentation.
- To assess the efficacy of DenseNet201, EfficientNet-B0, Mobilenet-v2, ResNet101, and Xception for coronary artery stenosis classification.
Main Methods:
- Coronary artery segmentation was performed using U-Net, ResUNet-a, and UNet++.
- Stenosis classification was conducted using DenseNet201, EfficientNet-B0, Mobilenet-v2, ResNet101, and Xception.
- Model performance was evaluated using metrics such as Dice score, Jaccard Index, accuracy, specificity, PPV, Cohen's Kappa, and AUC.
Main Results:
- U-Net achieved the highest segmentation performance with a 0.8467 Dice score and 0.7454 Jaccard Index.
- DenseNet201 demonstrated superior classification performance with 0.9000 accuracy, 0.9833 specificity, 0.9556 PPV, 0.7746 Cohen's Kappa, and 0.9694 AUC.
Conclusions:
- Automated deep learning approaches significantly enhance the accuracy and efficiency of coronary artery stenosis detection.
- U-Net is a highly effective model for coronary artery segmentation.
- DenseNet201 shows strong potential for reliable classification of coronary artery stenosis, aiding clinical decision-making.
Abstract:
In recent years, the prevalence of coronary artery disease (CAD) has become one of the leading causes of death around the world. Accurate stenosis detection of coronary arteries is crucial for timely treatment. Cardiologists use visual estimations when reading coronary angiography images to diagnose stenosis. As a result, they face various challenges which include high workloads, long processing times and human error. Computer-aided segmentation and classification of coronary arteries, as to whether stenosis is present or not, significantly reduces the workload of cardiologists and human errors caused by manual processes. Moreover, deep learning techniques have been shown to aid medical experts in diagnosing diseases using biomedical imaging. Thus, this study proposes the use of automatic segmentation of coronary arteries using U-Net, ResUNet-a, UNet++, models and classification using DenseNet201, EfficientNet-B0, Mobilenet-v2, ResNet101 and Xception models. In the case of segmentation, the comparative analysis of the three models has shown that U-Net achieved the highest score with a 0.8467 Dice score and 0.7454 Jaccard Index in comparison with UNet++ and ResUnet-a. Evaluation of the classification model's performances has shown that DenseNet201 performed better than other pretrained models with 0.9000 accuracy, 0.9833 specificity, 0.9556 PPV, 0.7746 Cohen's Kappa and 0.9694 Area Under the Curve (AUC).
Related Concept Videos
01:28Imaging Studies for Cardiovascular System V: CT
01:30Coronary Artery Disease I: Introduction