Deep learning framework for prediction of infection severity of COVID-19
With the onset of the COVID-19 pandemic, quantifying the condition of positively diagnosed patients is of paramount importance. Chest CT scans can be used to measure the severity of a lung infection and the isolate involvement sites in order to increase awareness of a patient's disease progress...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-08-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2022.940960/full |
| _version_ | 1811215020596396032 |
|---|---|
| author | Mehdi Yousefzadeh Mehdi Yousefzadeh Masoud Hasanpour Mozhdeh Zolghadri Fatemeh Salimi Ava Yektaeian Vaziri Abolfazl Mahmoudi Aqeel Abadi Ramezan Jafari Parsa Esfahanian Mohammad-Reza Nazem-Zadeh Mohammad-Reza Nazem-Zadeh |
| author_facet | Mehdi Yousefzadeh Mehdi Yousefzadeh Masoud Hasanpour Mozhdeh Zolghadri Fatemeh Salimi Ava Yektaeian Vaziri Abolfazl Mahmoudi Aqeel Abadi Ramezan Jafari Parsa Esfahanian Mohammad-Reza Nazem-Zadeh Mohammad-Reza Nazem-Zadeh |
| author_sort | Mehdi Yousefzadeh |
| collection | DOAJ |
| description | With the onset of the COVID-19 pandemic, quantifying the condition of positively diagnosed patients is of paramount importance. Chest CT scans can be used to measure the severity of a lung infection and the isolate involvement sites in order to increase awareness of a patient's disease progression. In this work, we developed a deep learning framework for lung infection severity prediction. To this end, we collected a dataset of 232 chest CT scans and involved two public datasets with an additional 59 scans for our model's training and used two external test sets with 21 scans for evaluation. On an input chest Computer Tomography (CT) scan, our framework, in parallel, performs a lung lobe segmentation utilizing a pre-trained model and infection segmentation using three distinct trained SE-ResNet18 based U-Net models, one for each of the axial, coronal, and sagittal views. By having the lobe and infection segmentation masks, we calculate the infection severity percentage in each lobe and classify that percentage into 6 categories of infection severity score using a k-nearest neighbors (k-NN) model. The lobe segmentation model achieved a Dice Similarity Score (DSC) in the range of [0.918, 0.981] for different lung lobes and our infection segmentation models gained DSC scores of 0.7254 and 0.7105 on our two test sets, respectfully. Similarly, two resident radiologists were assigned the same infection segmentation tasks, for which they obtained a DSC score of 0.7281 and 0.6693 on the two test sets. At last, performance on infection severity score over the entire test datasets was calculated, for which the framework's resulted in a Mean Absolute Error (MAE) of 0.505 ± 0.029, while the resident radiologists' was 0.571 ± 0.039. |
| first_indexed | 2024-04-12T06:15:17Z |
| format | Article |
| id | doaj.art-cbc68e768855437fbf06aa4abd8dd416 |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-04-12T06:15:17Z |
| publishDate | 2022-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-cbc68e768855437fbf06aa4abd8dd4162022-12-22T03:44:32ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2022-08-01910.3389/fmed.2022.940960940960Deep learning framework for prediction of infection severity of COVID-19Mehdi Yousefzadeh0Mehdi Yousefzadeh1Masoud Hasanpour2Mozhdeh Zolghadri3Fatemeh Salimi4Ava Yektaeian Vaziri5Abolfazl Mahmoudi Aqeel Abadi6Ramezan Jafari7Parsa Esfahanian8Mohammad-Reza Nazem-Zadeh9Mohammad-Reza Nazem-Zadeh10Department of Physics, Shahid Beheshti University, Tehran, IranSchool of Computer Science, Institute for Research in Fundamental Sciences (IPM), Tehran, IranResearch Center for Molecular and Cellular Imaging, Tehran University of Medical Sciences, Tehran, IranDepartment of Medical Physics, School of Medicine, Mashhad University of Medical Sciences, Mashhad, IranResearch Center for Molecular and Cellular Imaging, Tehran University of Medical Sciences, Tehran, IranDepartment of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences (TUMS), Tehran, IranDepartment of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences (TUMS), Tehran, IranDepartment of Radiology, Health Research Center, Baqiyatallah University of Medical Sciences, Tehran, IranSchool of Computer Science, Institute for Research in Fundamental Sciences (IPM), Tehran, IranResearch Center for Molecular and Cellular Imaging, Tehran University of Medical Sciences, Tehran, IranDepartment of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences (TUMS), Tehran, IranWith the onset of the COVID-19 pandemic, quantifying the condition of positively diagnosed patients is of paramount importance. Chest CT scans can be used to measure the severity of a lung infection and the isolate involvement sites in order to increase awareness of a patient's disease progression. In this work, we developed a deep learning framework for lung infection severity prediction. To this end, we collected a dataset of 232 chest CT scans and involved two public datasets with an additional 59 scans for our model's training and used two external test sets with 21 scans for evaluation. On an input chest Computer Tomography (CT) scan, our framework, in parallel, performs a lung lobe segmentation utilizing a pre-trained model and infection segmentation using three distinct trained SE-ResNet18 based U-Net models, one for each of the axial, coronal, and sagittal views. By having the lobe and infection segmentation masks, we calculate the infection severity percentage in each lobe and classify that percentage into 6 categories of infection severity score using a k-nearest neighbors (k-NN) model. The lobe segmentation model achieved a Dice Similarity Score (DSC) in the range of [0.918, 0.981] for different lung lobes and our infection segmentation models gained DSC scores of 0.7254 and 0.7105 on our two test sets, respectfully. Similarly, two resident radiologists were assigned the same infection segmentation tasks, for which they obtained a DSC score of 0.7281 and 0.6693 on the two test sets. At last, performance on infection severity score over the entire test datasets was calculated, for which the framework's resulted in a Mean Absolute Error (MAE) of 0.505 ± 0.029, while the resident radiologists' was 0.571 ± 0.039.https://www.frontiersin.org/articles/10.3389/fmed.2022.940960/fullCOVID-19deep learninginfection segmentationlobe segmentationCT scanseverity score |
| spellingShingle | Mehdi Yousefzadeh Mehdi Yousefzadeh Masoud Hasanpour Mozhdeh Zolghadri Fatemeh Salimi Ava Yektaeian Vaziri Abolfazl Mahmoudi Aqeel Abadi Ramezan Jafari Parsa Esfahanian Mohammad-Reza Nazem-Zadeh Mohammad-Reza Nazem-Zadeh Deep learning framework for prediction of infection severity of COVID-19 Frontiers in Medicine COVID-19 deep learning infection segmentation lobe segmentation CT scan severity score |
| title | Deep learning framework for prediction of infection severity of COVID-19 |
| title_full | Deep learning framework for prediction of infection severity of COVID-19 |
| title_fullStr | Deep learning framework for prediction of infection severity of COVID-19 |
| title_full_unstemmed | Deep learning framework for prediction of infection severity of COVID-19 |
| title_short | Deep learning framework for prediction of infection severity of COVID-19 |
| title_sort | deep learning framework for prediction of infection severity of covid 19 |
| topic | COVID-19 deep learning infection segmentation lobe segmentation CT scan severity score |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2022.940960/full |
| work_keys_str_mv | AT mehdiyousefzadeh deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT mehdiyousefzadeh deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT masoudhasanpour deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT mozhdehzolghadri deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT fatemehsalimi deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT avayektaeianvaziri deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT abolfazlmahmoudiaqeelabadi deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT ramezanjafari deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT parsaesfahanian deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT mohammadrezanazemzadeh deeplearningframeworkforpredictionofinfectionseverityofcovid19 AT mohammadrezanazemzadeh deeplearningframeworkforpredictionofinfectionseverityofcovid19 |