Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-Rays
We demonstrate use of iteratively pruned deep learning model ensembles for detecting pulmonary manifestations of COVID-19 with chest X-rays. This disease is caused by the novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) virus, also known as the novel Coronavirus (2019-nCoV). A cust...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9121222/ |
| _version_ | 1818330559631327232 |
|---|---|
| author | Sivaramakrishnan Rajaraman Jenifer Siegelman Philip O. Alderson Lucas S. Folio Les R. Folio Sameer K. Antani |
| author_facet | Sivaramakrishnan Rajaraman Jenifer Siegelman Philip O. Alderson Lucas S. Folio Les R. Folio Sameer K. Antani |
| author_sort | Sivaramakrishnan Rajaraman |
| collection | DOAJ |
| description | We demonstrate use of iteratively pruned deep learning model ensembles for detecting pulmonary manifestations of COVID-19 with chest X-rays. This disease is caused by the novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) virus, also known as the novel Coronavirus (2019-nCoV). A custom convolutional neural network and a selection of ImageNet pretrained models are trained and evaluated at patient-level on publicly available CXR collections to learn modality-specific feature representations. The learned knowledge is transferred and fine-tuned to improve performance and generalization in the related task of classifying CXRs as normal, showing bacterial pneumonia, or COVID-19-viral abnormalities. The best performing models are iteratively pruned to reduce complexity and improve memory efficiency. The predictions of the best-performing pruned models are combined through different ensemble strategies to improve classification performance. Empirical evaluations demonstrate that the weighted average of the best-performing pruned models significantly improves performance resulting in an accuracy of 99.01% and area under the curve of 0.9972 in detecting COVID-19 findings on CXRs. The combined use of modality-specific knowledge transfer, iterative model pruning, and ensemble learning resulted in improved predictions. We expect that this model can be quickly adopted for COVID-19 screening using chest radiographs. |
| first_indexed | 2024-12-13T13:05:53Z |
| format | Article |
| id | doaj.art-d6f4898092bd482483bcebc63f8a57ff |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-13T13:05:53Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-d6f4898092bd482483bcebc63f8a57ff2022-12-21T23:44:50ZengIEEEIEEE Access2169-35362020-01-01811504111505010.1109/ACCESS.2020.30038109121222Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-RaysSivaramakrishnan Rajaraman0https://orcid.org/0000-0003-0871-8634Jenifer Siegelman1Philip O. Alderson2Lucas S. Folio3Les R. Folio4Sameer K. Antani5Lister Hill National Center for Biomedical Communications, National Library of Medicine, Bethesda, MD, USATakeda Pharmaceuticals, Cambridge, MA, USASchool of Medicine, Saint Louis University, St. Louis, MO, USAFunctional and Applied Biomechanics Section, Clinical Center, National Institutes of Health, Bethesda, MD, USARadiological and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, MD, USALister Hill National Center for Biomedical Communications, National Library of Medicine, Bethesda, MD, USAWe demonstrate use of iteratively pruned deep learning model ensembles for detecting pulmonary manifestations of COVID-19 with chest X-rays. This disease is caused by the novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) virus, also known as the novel Coronavirus (2019-nCoV). A custom convolutional neural network and a selection of ImageNet pretrained models are trained and evaluated at patient-level on publicly available CXR collections to learn modality-specific feature representations. The learned knowledge is transferred and fine-tuned to improve performance and generalization in the related task of classifying CXRs as normal, showing bacterial pneumonia, or COVID-19-viral abnormalities. The best performing models are iteratively pruned to reduce complexity and improve memory efficiency. The predictions of the best-performing pruned models are combined through different ensemble strategies to improve classification performance. Empirical evaluations demonstrate that the weighted average of the best-performing pruned models significantly improves performance resulting in an accuracy of 99.01% and area under the curve of 0.9972 in detecting COVID-19 findings on CXRs. The combined use of modality-specific knowledge transfer, iterative model pruning, and ensemble learning resulted in improved predictions. We expect that this model can be quickly adopted for COVID-19 screening using chest radiographs.https://ieeexplore.ieee.org/document/9121222/COVID-19convolutional neural networkdeep learningensembleiterative pruning |
| spellingShingle | Sivaramakrishnan Rajaraman Jenifer Siegelman Philip O. Alderson Lucas S. Folio Les R. Folio Sameer K. Antani Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-Rays IEEE Access COVID-19 convolutional neural network deep learning ensemble iterative pruning |
| title | Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-Rays |
| title_full | Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-Rays |
| title_fullStr | Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-Rays |
| title_full_unstemmed | Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-Rays |
| title_short | Iteratively Pruned Deep Learning Ensembles for COVID-19 Detection in Chest X-Rays |
| title_sort | iteratively pruned deep learning ensembles for covid 19 detection in chest x rays |
| topic | COVID-19 convolutional neural network deep learning ensemble iterative pruning |
| url | https://ieeexplore.ieee.org/document/9121222/ |
| work_keys_str_mv | AT sivaramakrishnanrajaraman iterativelypruneddeeplearningensemblesforcovid19detectioninchestxrays AT jenifersiegelman iterativelypruneddeeplearningensemblesforcovid19detectioninchestxrays AT philipoalderson iterativelypruneddeeplearningensemblesforcovid19detectioninchestxrays AT lucassfolio iterativelypruneddeeplearningensemblesforcovid19detectioninchestxrays AT lesrfolio iterativelypruneddeeplearningensemblesforcovid19detectioninchestxrays AT sameerkantani iterativelypruneddeeplearningensemblesforcovid19detectioninchestxrays |