COVID-19 prognostic modeling using CT radiomic features and machine learning algorithms: Analysis of a multi-institutional dataset of 14,339 patients

• Published in: Computers in biology and medicine Vol. 145; p. 105467
• Main Authors: Shiri, Isaac, Salimi, Yazdan, Pakbin, Masoumeh, Hajianfar, Ghasem, Avval, Atlas Haddadi, Sanaat, Amirhossein, Mostafaei, Shayan, Akhavanallaf, Azadeh, Saberi, Abdollah, Mansouri, Zahra, Askari, Dariush, Ghasemian, Mohammadreza, Sharifipour, Ehsan, Sandoughdaran, Saleh, Sohrabi, Ahmad, Sadati, Elham, Livani, Somayeh, Iranpour, Pooya, Kolahi, Shahriar, Khateri, Maziar, Bijari, Salar, Atashzar, Mohammad Reza, Shayesteh, Sajad P., Khosravi, Bardia, Babaei, Mohammad Reza, Jenabi, Elnaz, Hasanian, Mohammad, Shahhamzeh, Alireza, Foroghi Ghomi, Seyaed Yaser, Mozafari, Abolfazl, Teimouri, Arash, Movaseghi, Fatemeh, Ahmari, Azin, Goharpey, Neda, Bozorgmehr, Rama, Shirzad-Aski, Hesamaddin, Mortazavi, Roozbeh, Karimi, Jalal, Mortazavi, Nazanin, Besharat, Sima, Afsharpad, Mandana, Abdollahi, Hamid, Geramifar, Parham, Radmard, Amir Reza, Arabi, Hossein, Rezaei-Kalantari, Kiara, Oveisi, Mehrdad, Rahmim, Arman, Zaidi, Habib
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.06.2022; Elsevier Limited; The Authors. Published by Elsevier Ltd
• Subjects: Algorithms; Artificial intelligence; Classifiers; Coronaviruses; COVID-19; COVID-19 - diagnostic imaging; Datasets; Deep learning; Feature extraction; Humans; Internal Medicine; Lung Neoplasms; Lungs; Machine Learning; Medical prognosis; Modelling; Other; Patients; Polymerase chain reaction; Prognosis; Radiomics; Retrospective Studies; Sensitivity; Statistical analysis; Tomography, X-Ray Computed - methods; Variance analysis; X-ray CT; COVID-19; X-ray CT; Prognosis; Machine learning; Radiomics
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2022.105467

We aimed to analyze the prognostic power of CT-based radiomics models using data of 14,339 COVID-19 patients. Whole lung segmentations were performed automatically using a deep learning-based model to extract 107 intensity and texture radiomics features. We used four feature selection algorithms and seven classifiers. We evaluated the models using ten different splitting and cross-validation strategies, including non-harmonized and ComBat-harmonized datasets. The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were reported. In the test dataset (4,301) consisting of CT and/or RT-PCR positive cases, AUC, sensitivity, and specificity of 0.83 ± 0.01 (CI95%: 0.81–0.85), 0.81, and 0.72, respectively, were obtained by ANOVA feature selector + Random Forest (RF) classifier. Similar results were achieved in RT-PCR-only positive test sets (3,644). In ComBat harmonized dataset, Relief feature selector + RF classifier resulted in the highest performance of AUC, reaching 0.83 ± 0.01 (CI95%: 0.81–0.85), with a sensitivity and specificity of 0.77 and 0.74, respectively. ComBat harmonization did not depict statistically significant improvement compared to a non-harmonized dataset. In leave-one-center-out, the combination of ANOVA feature selector and RF classifier resulted in the highest performance. Lung CT radiomics features can be used for robust prognostic modeling of COVID-19. The predictive power of the proposed CT radiomics model is more reliable when using a large multicentric heterogeneous dataset, and may be used prospectively in clinical setting to manage COVID-19 patients. [Display omitted] •CT images of 14′339 COVID-19 patients with known outcome from 19 centers were enrolled.•28 combinations of feature selection and classification approaches were implemented.•The models were evaluated using 10 different splitting and cross-validation strategies.•Lung CT radiomics features are promising for generalizable prognostic modeling.