The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries
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...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/13/13/2274 |
| _version_ | 1797591914942300160 |
|---|---|
| author | Şerife Kaba Huseyin Haci Ali Isin Ahmet Ilhan Cenk Conkbayir |
| author_facet | Şerife Kaba Huseyin Haci Ali Isin Ahmet Ilhan Cenk Conkbayir |
| author_sort | Şerife Kaba |
| collection | DOAJ |
| description | 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). |
| first_indexed | 2024-03-11T01:44:12Z |
| format | Article |
| id | doaj.art-a0d507f2fd79455ba1982987c9598d85 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-11T01:44:12Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-a0d507f2fd79455ba1982987c9598d852023-11-18T16:22:26ZengMDPI AGDiagnostics2075-44182023-07-011313227410.3390/diagnostics13132274The Application of Deep Learning for the Segmentation and Classification of Coronary ArteriesŞerife Kaba0Huseyin Haci1Ali Isin2Ahmet Ilhan3Cenk Conkbayir4Department of Biomedical Engineering, Near East University, TRNC Mersin 10, Nicosia 99138, TurkeyDepartment of Electrical-Electronic Engineering, Near East University, TRNC Mersin 10, Nicosia 99138, TurkeyDepartment of Biomedical Engineering, Cyprus International University, TRNC Mersin 10, Nicosia 99138, TurkeyDepartment of Computer Engineering, Near East University, TRNC Mersin 10, Nicosia 99138, TurkeyDepartment of Cardiology, Near East University, TRNC Mersin 10, Nicosia 99138, TurkeyIn 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).https://www.mdpi.com/2075-4418/13/13/2274coronary artery disease (CAD)coronary arteriesangiographyU-Netpretrained models |
| spellingShingle | Şerife Kaba Huseyin Haci Ali Isin Ahmet Ilhan Cenk Conkbayir The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries Diagnostics coronary artery disease (CAD) coronary arteries angiography U-Net pretrained models |
| title | The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries |
| title_full | The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries |
| title_fullStr | The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries |
| title_full_unstemmed | The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries |
| title_short | The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries |
| title_sort | application of deep learning for the segmentation and classification of coronary arteries |
| topic | coronary artery disease (CAD) coronary arteries angiography U-Net pretrained models |
| url | https://www.mdpi.com/2075-4418/13/13/2274 |
| work_keys_str_mv | AT serifekaba theapplicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT huseyinhaci theapplicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT aliisin theapplicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT ahmetilhan theapplicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT cenkconkbayir theapplicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT serifekaba applicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT huseyinhaci applicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT aliisin applicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT ahmetilhan applicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries AT cenkconkbayir applicationofdeeplearningforthesegmentationandclassificationofcoronaryarteries |