COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancing
The novel COVID-19 is a global pandemic disease overgrowing worldwide. Computer-aided screening tools with greater sensitivity are imperative for disease diagnosis and prognosis as early as possible. It also can be a helpful tool in triage for testing and clinical supervision of COVID-19 patients. H...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-01-01
|
| Series: | Informatics in Medicine Unlocked |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S235291482100191X |
| _version_ | 1818824807150518272 |
|---|---|
| author | Md. Kamrul Hasan Md. Tasnim Jawad Kazi Nasim Imtiaz Hasan Sajal Basak Partha Md. Masum Al Masba Shumit Saha Mohammad Ali Moni |
| author_facet | Md. Kamrul Hasan Md. Tasnim Jawad Kazi Nasim Imtiaz Hasan Sajal Basak Partha Md. Masum Al Masba Shumit Saha Mohammad Ali Moni |
| author_sort | Md. Kamrul Hasan |
| collection | DOAJ |
| description | The novel COVID-19 is a global pandemic disease overgrowing worldwide. Computer-aided screening tools with greater sensitivity are imperative for disease diagnosis and prognosis as early as possible. It also can be a helpful tool in triage for testing and clinical supervision of COVID-19 patients. However, designing such an automated tool from non-invasive radiographic images is challenging as many manually annotated datasets are not publicly available yet, which is the essential core requirement of supervised learning schemes. This article proposes a 3D Convolutional Neural Network (CNN)-based classification approach considering both the inter-and intra-slice spatial voxel information. The proposed system is trained end-to-end on the 3D patches from the whole volumetric Computed Tomography (CT) images to enlarge the number of training samples, performing the ablation studies on patch size determination. We integrate progressive resizing, segmentation, augmentations, and class-rebalancing into our 3D network. The segmentation is a critical prerequisite step for COVID-19 diagnosis enabling the classifier to learn prominent lung features while excluding the outer lung regions of the CT scans. We evaluate all the extensive experiments on a publicly available dataset named MosMed, having binary- and multi-class chest CT image partitions. Our experimental results are very encouraging, yielding areas under the Receiver Operating Characteristics (ROC) curve of 0.914±0.049and 0.893±0.035for the binary- and multi-class tasks, respectively, applying 5-fold cross-validations. Our method’s promising results delegate it as a favorable aiding tool for clinical practitioners and radiologists to assess COVID-19. |
| first_indexed | 2024-12-19T00:01:44Z |
| format | Article |
| id | doaj.art-10df2f8670c14a229fc98a29342d09d1 |
| institution | Directory Open Access Journal |
| issn | 2352-9148 |
| language | English |
| last_indexed | 2024-12-19T00:01:44Z |
| publishDate | 2021-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Informatics in Medicine Unlocked |
| spelling | doaj.art-10df2f8670c14a229fc98a29342d09d12022-12-21T20:46:26ZengElsevierInformatics in Medicine Unlocked2352-91482021-01-0126100709COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancingMd. Kamrul Hasan0Md. Tasnim Jawad1Kazi Nasim Imtiaz Hasan2Sajal Basak Partha3Md. Masum Al Masba4Shumit Saha5Mohammad Ali Moni6Department of Electrical and Electronic Engineering, Khulna University of Engineering & Technology, Khulna-9203, Bangladesh; Correspondence to: Department of EEE, KUET, Khulna-9203, Bangladesh.Department of Electrical and Electronic Engineering, Khulna University of Engineering & Technology, Khulna-9203, BangladeshDepartment of Computer Science and Engineering, Khulna University of Engineering & Technology, Khulna-9203, BangladeshDepartment of Computer Science and Engineering, Khulna University of Engineering & Technology, Khulna-9203, BangladeshDepartment of Computer Science and Engineering, Khulna University of Engineering & Technology, Khulna-9203, BangladeshDepartment of Electronics and Communication Engineering, Khulna University of Engineering & Technology, Khulna-9203, BangladeshDepartment of Computer Science & Engineering, Pabna University of Science and Technology, Pabna-6600, Bangladesh; School of Health and Rehabilitation Sciences, Faculty of Health and Behavioural Sciences, The University of Queensland, St Lucia, QLD 4072, AustraliaThe novel COVID-19 is a global pandemic disease overgrowing worldwide. Computer-aided screening tools with greater sensitivity are imperative for disease diagnosis and prognosis as early as possible. It also can be a helpful tool in triage for testing and clinical supervision of COVID-19 patients. However, designing such an automated tool from non-invasive radiographic images is challenging as many manually annotated datasets are not publicly available yet, which is the essential core requirement of supervised learning schemes. This article proposes a 3D Convolutional Neural Network (CNN)-based classification approach considering both the inter-and intra-slice spatial voxel information. The proposed system is trained end-to-end on the 3D patches from the whole volumetric Computed Tomography (CT) images to enlarge the number of training samples, performing the ablation studies on patch size determination. We integrate progressive resizing, segmentation, augmentations, and class-rebalancing into our 3D network. The segmentation is a critical prerequisite step for COVID-19 diagnosis enabling the classifier to learn prominent lung features while excluding the outer lung regions of the CT scans. We evaluate all the extensive experiments on a publicly available dataset named MosMed, having binary- and multi-class chest CT image partitions. Our experimental results are very encouraging, yielding areas under the Receiver Operating Characteristics (ROC) curve of 0.914±0.049and 0.893±0.035for the binary- and multi-class tasks, respectively, applying 5-fold cross-validations. Our method’s promising results delegate it as a favorable aiding tool for clinical practitioners and radiologists to assess COVID-19.http://www.sciencedirect.com/science/article/pii/S235291482100191XCOVID-193D convolutional neural networkVolumetric chest CT scans3D patchesProgressive resizing |
| spellingShingle | Md. Kamrul Hasan Md. Tasnim Jawad Kazi Nasim Imtiaz Hasan Sajal Basak Partha Md. Masum Al Masba Shumit Saha Mohammad Ali Moni COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancing Informatics in Medicine Unlocked COVID-19 3D convolutional neural network Volumetric chest CT scans 3D patches Progressive resizing |
| title | COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancing |
| title_full | COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancing |
| title_fullStr | COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancing |
| title_full_unstemmed | COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancing |
| title_short | COVID-19 identification from volumetric chest CT scans using a progressively resized 3D-CNN incorporating segmentation, augmentation, and class-rebalancing |
| title_sort | covid 19 identification from volumetric chest ct scans using a progressively resized 3d cnn incorporating segmentation augmentation and class rebalancing |
| topic | COVID-19 3D convolutional neural network Volumetric chest CT scans 3D patches Progressive resizing |
| url | http://www.sciencedirect.com/science/article/pii/S235291482100191X |
| work_keys_str_mv | AT mdkamrulhasan covid19identificationfromvolumetricchestctscansusingaprogressivelyresized3dcnnincorporatingsegmentationaugmentationandclassrebalancing AT mdtasnimjawad covid19identificationfromvolumetricchestctscansusingaprogressivelyresized3dcnnincorporatingsegmentationaugmentationandclassrebalancing AT kazinasimimtiazhasan covid19identificationfromvolumetricchestctscansusingaprogressivelyresized3dcnnincorporatingsegmentationaugmentationandclassrebalancing AT sajalbasakpartha covid19identificationfromvolumetricchestctscansusingaprogressivelyresized3dcnnincorporatingsegmentationaugmentationandclassrebalancing AT mdmasumalmasba covid19identificationfromvolumetricchestctscansusingaprogressivelyresized3dcnnincorporatingsegmentationaugmentationandclassrebalancing AT shumitsaha covid19identificationfromvolumetricchestctscansusingaprogressivelyresized3dcnnincorporatingsegmentationaugmentationandclassrebalancing AT mohammadalimoni covid19identificationfromvolumetricchestctscansusingaprogressivelyresized3dcnnincorporatingsegmentationaugmentationandclassrebalancing |