Machine learning-based prognostic modeling using clinical data and quantitative radiomic features from chest CT images in COVID-19 patients

Shiri, I. and Sorouri, M. and Geramifar, P. and Nazari, M. and Abdollahi, M. and Salimi, Y. and Khosravi, B. and Askari, D. and Aghaghazvini, L. and Hajianfar, G. and Kasaeian, A. and Abdollahi, H. and Arabi, H. and Rahmim, A. and Radmard, A.R. and Zaidi, H. (2021) Machine learning-based prognostic modeling using clinical data and quantitative radiomic features from chest CT images in COVID-19 patients. Computers in Biology and Medicine, 132.

[img] Text
Machine learning-based prognostic modeling using clinical data and quantitative radiomic features from chest CT images in COVID-19 patients.pdf

Download (7MB)
Official URL:


Objective: To develop prognostic models for survival (alive or deceased status) prediction of COVID-19 patients using clinical data (demographics and history, laboratory tests, visual scoring by radiologists) and lung/lesion radiomic features extracted from chest CT images. Methods: Overall, 152 patients were enrolled in this study protocol. These were divided into 106 training/validation and 46 test datasets (untouched during training), respectively. Radiomic features were extracted from the segmented lungs and infectious lesions separately from chest CT images. Clinical data, including patients� history and demographics, laboratory tests and radiological scores were also collected. Univariate analysis was first performed (q-value reported after false discovery rate (FDR) correction) to determine the most predictive features among all imaging and clinical data. Prognostic modeling of survival was performed using radiomic features and clinical data, separately or in combination. Maximum relevance minimum redundancy (MRMR) and XGBoost were used for feature selection and classification. The receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC), sensitivity, specificity, and accuracy were used to assess the prognostic performance of the models on the test datasets. Results: For clinical data, cancer comorbidity (q-value < 0.01), consciousness level (q-value < 0.05) and radiological score involved zone (q-value < 0.02) were found to have high correlated features with outcome. Oxygen saturation (AUC = 0.73, q-value < 0.01) and Blood Urea Nitrogen (AUC = 0.72, q-value = 0.72) were identified as high clinical features. For lung radiomic features, SAHGLE (AUC = 0.70) and HGLZE (AUC = 0.67) from GLSZM were identified as most prognostic features. Amongst lesion radiomic features, RLNU from GLRLM (AUC = 0.73), HGLZE from GLSZM (AUC = 0.73) had the highest performance. In multivariate analysis, combining lung, lesion and clinical features was determined to provide the most accurate prognostic model (AUC = 0.95 ± 0.029 (95CI: 0.95�0.96), accuracy = 0.88 ± 0.046 (95 CI: 0.88�0.89), sensitivity = 0.88 ± 0.066 (95 CI = 0.87�0.9) and specificity = 0.89 ± 0.07 (95 CI = 0.87�0.9)). Conclusion: Combination of radiomic features and clinical data can effectively predict outcome in COVID-19 patients. The developed model has significant potential for improved management of COVID-19 patients. © 2021 The Author(s)

Item Type: Article
Additional Information: cited By 4
Uncontrolled Keywords: Biological organs; Diagnosis; Machine learning; Multivariant analysis; Population statistics; Urea, Area under the ROC curve; Clinical data; Computed tomography; Computed tomography images; COVID-19; Modeling; Prognose; Prognostic model; Q-values; Radiomic, Computerized tomography, hydroxychloroquine; lopinavir plus ritonavir, adult; aged; Article; clinical feature; clinical study; comorbidity; computer assisted tomography; consciousness level; controlled study; coronavirus disease 2019; feature selection; female; human; human cell; image segmentation; laboratory test; lung lesion; machine learning; major clinical study; male; oxygen saturation; priority journal; radiomics; receiver operating characteristic; retrospective study; sensitivity and specificity; univariate analysis; urea nitrogen blood level; machine learning; prognosis; x-ray computed tomography, COVID-19; Humans; Machine Learning; Prognosis; Retrospective Studies; SARS-CoV-2; Tomography, X-Ray Computed
Subjects: WC Communicable Diseases
WN Radiology. Diagnostic Imaging
Depositing User: eprints admin
Date Deposited: 02 Jan 2022 09:03
Last Modified: 02 Jan 2022 09:03

Actions (login required)

View Item View Item