Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual Evaluation
<b>Background/Objectives:</b> To develop a computer-aided diagnosis (CAD) method for the classification of late gadolinium enhancement (LGE) cardiac MRI images into myocardial infarction (MI), myocarditis, and healthy classes using a fine-tuned VGG16 model hybridized with multi-layer per...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/2/207 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588721956847616 |
|---|---|
| author | Amani Ben Khalifa Manel Mili Mezri Maatouk Asma Ben Abdallah Mabrouk Abdellali Sofiene Gaied Azza Ben Ali Yassir Lahouel Mohamed Hedi Bedoui Ahmed Zrig |
| author_facet | Amani Ben Khalifa Manel Mili Mezri Maatouk Asma Ben Abdallah Mabrouk Abdellali Sofiene Gaied Azza Ben Ali Yassir Lahouel Mohamed Hedi Bedoui Ahmed Zrig |
| author_sort | Amani Ben Khalifa |
| collection | DOAJ |
| description | <b>Background/Objectives:</b> To develop a computer-aided diagnosis (CAD) method for the classification of late gadolinium enhancement (LGE) cardiac MRI images into myocardial infarction (MI), myocarditis, and healthy classes using a fine-tuned VGG16 model hybridized with multi-layer perceptron (MLP) (VGG16-MLP) and assess our model’s performance in comparison to various pre-trained base models and MRI readers. <b>Methods:</b> This study included 361 LGE images for MI, 222 for myocarditis, and 254 for the healthy class. The left ventricle was extracted automatically using a U-net segmentation model on LGE images. Fine-tuned VGG16 was performed for feature extraction. A spatial attention mechanism was implemented as a part of the neural network architecture. The MLP architecture was used for the classification. The evaluation metrics were calculated using a separate test set. To compare the VGG16 model’s performance in feature extraction, various pre-trained base models were evaluated: VGG19, DenseNet121, DenseNet201, MobileNet, InceptionV3, and InceptionResNetV2. The Support Vector Machine (SVM) classifier was evaluated and compared to MLP for the classification task. The performance of the VGG16-MLP model was compared with a subjective visual analysis conducted by two blinded independent readers. <b>Results:</b> The VGG16-MLP model allowed high-performance differentiation between MI, myocarditis, and healthy LGE cardiac MRI images. It outperformed the other tested models with 96% accuracy, 97% precision, 96% sensitivity, and 96% F1-score. Our model surpassed the accuracy of Reader 1 by 27% and Reader 2 by 17%. <b>Conclusions:</b> Our study demonstrated that the VGG16-MLP model permits accurate classification of MI, myocarditis, and healthy LGE cardiac MRI images and could be considered a reliable computer-aided diagnosis approach specifically for radiologists with limited experience in cardiovascular imaging. |
| format | Article |
| id | doaj-art-bda148097b0d436095044dda055c6522 |
| institution | Kabale University |
| issn | 2075-4418 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-bda148097b0d436095044dda055c65222025-01-24T13:29:06ZengMDPI AGDiagnostics2075-44182025-01-0115220710.3390/diagnostics15020207Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual EvaluationAmani Ben Khalifa0Manel Mili1Mezri Maatouk2Asma Ben Abdallah3Mabrouk Abdellali4Sofiene Gaied5Azza Ben Ali6Yassir Lahouel7Mohamed Hedi Bedoui8Ahmed Zrig9Technology and Medical Imaging Laboratory LR12ES06, Faculty of Medicine of Monastir, University of Monastir, Monastir 5019, TunisiaTechnology and Medical Imaging Laboratory LR12ES06, Faculty of Medicine of Monastir, University of Monastir, Monastir 5019, TunisiaLR18-SP08 Department of Radiology, University Hospital of Monastir, Monastir 5019, TunisiaTechnology and Medical Imaging Laboratory LR12ES06, Faculty of Medicine of Monastir, University of Monastir, Monastir 5019, TunisiaLR18-SP08 Department of Radiology, University Hospital of Monastir, Monastir 5019, TunisiaLR18-SP08 Department of Radiology, University Hospital of Monastir, Monastir 5019, TunisiaLR18-SP08 Department of Radiology, University Hospital of Monastir, Monastir 5019, TunisiaLR18-SP08 Department of Radiology, University Hospital of Monastir, Monastir 5019, TunisiaTechnology and Medical Imaging Laboratory LR12ES06, Faculty of Medicine of Monastir, University of Monastir, Monastir 5019, TunisiaLR18-SP08 Department of Radiology, University Hospital of Monastir, Monastir 5019, Tunisia<b>Background/Objectives:</b> To develop a computer-aided diagnosis (CAD) method for the classification of late gadolinium enhancement (LGE) cardiac MRI images into myocardial infarction (MI), myocarditis, and healthy classes using a fine-tuned VGG16 model hybridized with multi-layer perceptron (MLP) (VGG16-MLP) and assess our model’s performance in comparison to various pre-trained base models and MRI readers. <b>Methods:</b> This study included 361 LGE images for MI, 222 for myocarditis, and 254 for the healthy class. The left ventricle was extracted automatically using a U-net segmentation model on LGE images. Fine-tuned VGG16 was performed for feature extraction. A spatial attention mechanism was implemented as a part of the neural network architecture. The MLP architecture was used for the classification. The evaluation metrics were calculated using a separate test set. To compare the VGG16 model’s performance in feature extraction, various pre-trained base models were evaluated: VGG19, DenseNet121, DenseNet201, MobileNet, InceptionV3, and InceptionResNetV2. The Support Vector Machine (SVM) classifier was evaluated and compared to MLP for the classification task. The performance of the VGG16-MLP model was compared with a subjective visual analysis conducted by two blinded independent readers. <b>Results:</b> The VGG16-MLP model allowed high-performance differentiation between MI, myocarditis, and healthy LGE cardiac MRI images. It outperformed the other tested models with 96% accuracy, 97% precision, 96% sensitivity, and 96% F1-score. Our model surpassed the accuracy of Reader 1 by 27% and Reader 2 by 17%. <b>Conclusions:</b> Our study demonstrated that the VGG16-MLP model permits accurate classification of MI, myocarditis, and healthy LGE cardiac MRI images and could be considered a reliable computer-aided diagnosis approach specifically for radiologists with limited experience in cardiovascular imaging.https://www.mdpi.com/2075-4418/15/2/207deep learningVGG16myocardial infarctionmyocarditis |
| spellingShingle | Amani Ben Khalifa Manel Mili Mezri Maatouk Asma Ben Abdallah Mabrouk Abdellali Sofiene Gaied Azza Ben Ali Yassir Lahouel Mohamed Hedi Bedoui Ahmed Zrig Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual Evaluation Diagnostics deep learning VGG16 myocardial infarction myocarditis |
| title | Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual Evaluation |
| title_full | Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual Evaluation |
| title_fullStr | Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual Evaluation |
| title_full_unstemmed | Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual Evaluation |
| title_short | Deep Transfer Learning for Classification of Late Gadolinium Enhancement Cardiac MRI Images into Myocardial Infarction, Myocarditis, and Healthy Classes: Comparison with Subjective Visual Evaluation |
| title_sort | deep transfer learning for classification of late gadolinium enhancement cardiac mri images into myocardial infarction myocarditis and healthy classes comparison with subjective visual evaluation |
| topic | deep learning VGG16 myocardial infarction myocarditis |
| url | https://www.mdpi.com/2075-4418/15/2/207 |
| work_keys_str_mv | AT amanibenkhalifa deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT manelmili deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT mezrimaatouk deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT asmabenabdallah deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT mabroukabdellali deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT sofienegaied deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT azzabenali deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT yassirlahouel deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT mohamedhedibedoui deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation AT ahmedzrig deeptransferlearningforclassificationoflategadoliniumenhancementcardiacmriimagesintomyocardialinfarctionmyocarditisandhealthyclassescomparisonwithsubjectivevisualevaluation |