Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray Images
Thorax disease is a life-threatening disease caused by bacterial infections that occur in the lungs. It could be deadly if not treated at the right time, so early diagnosis of thoracic diseases is vital. The suggested study can assist radiologists in more swiftly diagnosing thorax disorders and in t...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/13/22/3462 |
| _version_ | 1827640158666620928 |
|---|---|
| author | Zeeshan Ahmad Ahmad Kamran Malik Nafees Qamar Saif ul Islam |
| author_facet | Zeeshan Ahmad Ahmad Kamran Malik Nafees Qamar Saif ul Islam |
| author_sort | Zeeshan Ahmad |
| collection | DOAJ |
| description | Thorax disease is a life-threatening disease caused by bacterial infections that occur in the lungs. It could be deadly if not treated at the right time, so early diagnosis of thoracic diseases is vital. The suggested study can assist radiologists in more swiftly diagnosing thorax disorders and in the rapid airport screening of patients with a thorax disease, such as pneumonia. This paper focuses on automatically detecting and localizing thorax disease using chest X-ray images. It provides accurate detection and localization using DenseNet-121 which is foundation of our proposed framework, called Z-Net. The proposed framework utilizes the weighted cross-entropy loss function (W-CEL) that manages class imbalance issue in the ChestX-ray14 dataset, which helped in achieving the highest performance as compared to the previous models. The 112,120 images contained in the ChestX-ray14 dataset (60,412 images are normal, and the rest contain thorax diseases) were preprocessed and then trained for classification and localization. This work uses computer-aided diagnosis (CAD) system that supports development of highly accurate and precise computer-aided systems. We aim to develop a CAD system using a deep learning approach. Our quantitative results show high AUC scores in comparison with the latest research works. The proposed approach achieved the highest mean AUC score of 85.8%. This is the highest accuracy documented in the literature for any related model. |
| first_indexed | 2024-03-09T16:54:01Z |
| format | Article |
| id | doaj.art-a60487ea72614012b7731e4aae606438 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-09T16:54:01Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-a60487ea72614012b7731e4aae6064382023-11-24T14:37:42ZengMDPI AGDiagnostics2075-44182023-11-011322346210.3390/diagnostics13223462Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray ImagesZeeshan Ahmad0Ahmad Kamran Malik1Nafees Qamar2Saif ul Islam3Department of Computer Science, COMSATS University Islamabad, Islamabad 45550, PakistanDepartment of Computer Science, COMSATS University Islamabad, Islamabad 45550, PakistanSchool of Health and Behavioral Sciences, Bryant University, Smithfield, RI 02917, USADepartment of Computer Science, Institute of Space Technology, Islamabad 44000, PakistanThorax disease is a life-threatening disease caused by bacterial infections that occur in the lungs. It could be deadly if not treated at the right time, so early diagnosis of thoracic diseases is vital. The suggested study can assist radiologists in more swiftly diagnosing thorax disorders and in the rapid airport screening of patients with a thorax disease, such as pneumonia. This paper focuses on automatically detecting and localizing thorax disease using chest X-ray images. It provides accurate detection and localization using DenseNet-121 which is foundation of our proposed framework, called Z-Net. The proposed framework utilizes the weighted cross-entropy loss function (W-CEL) that manages class imbalance issue in the ChestX-ray14 dataset, which helped in achieving the highest performance as compared to the previous models. The 112,120 images contained in the ChestX-ray14 dataset (60,412 images are normal, and the rest contain thorax diseases) were preprocessed and then trained for classification and localization. This work uses computer-aided diagnosis (CAD) system that supports development of highly accurate and precise computer-aided systems. We aim to develop a CAD system using a deep learning approach. Our quantitative results show high AUC scores in comparison with the latest research works. The proposed approach achieved the highest mean AUC score of 85.8%. This is the highest accuracy documented in the literature for any related model.https://www.mdpi.com/2075-4418/13/22/3462thorax diseasechest X-rayDeep Convolutional Neural Network (DCNN)image processingclassification |
| spellingShingle | Zeeshan Ahmad Ahmad Kamran Malik Nafees Qamar Saif ul Islam Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray Images Diagnostics thorax disease chest X-ray Deep Convolutional Neural Network (DCNN) image processing classification |
| title | Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray Images |
| title_full | Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray Images |
| title_fullStr | Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray Images |
| title_full_unstemmed | Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray Images |
| title_short | Efficient Thorax Disease Classification and Localization Using DCNN and Chest X-ray Images |
| title_sort | efficient thorax disease classification and localization using dcnn and chest x ray images |
| topic | thorax disease chest X-ray Deep Convolutional Neural Network (DCNN) image processing classification |
| url | https://www.mdpi.com/2075-4418/13/22/3462 |
| work_keys_str_mv | AT zeeshanahmad efficientthoraxdiseaseclassificationandlocalizationusingdcnnandchestxrayimages AT ahmadkamranmalik efficientthoraxdiseaseclassificationandlocalizationusingdcnnandchestxrayimages AT nafeesqamar efficientthoraxdiseaseclassificationandlocalizationusingdcnnandchestxrayimages AT saifulislam efficientthoraxdiseaseclassificationandlocalizationusingdcnnandchestxrayimages |