COVID-19 prognostic modeling using CT radiomic features and machine learning algorithms: Analysis of a multi-institutional dataset of 14,339 patients
Isaac Shiri, Yazdan Salimi, Masoumeh Pakbin, Ghasem Hajianfar, Atlas Haddadi Avval, Amirhossein Sanaat, Shayan Mostafaei, Azadeh Akhavanallaf, Abdollah Saberi, Zahra Mansouri, Dariush Askari, Mohammadreza Ghasemian, Ehsan Sharifipour, Saleh Sandoughdaran, Ahmad Sohrabi, Elham Sadati, Somayeh Livani, Pooya Iranpour, Shahriar Kolahi, Maziar Khateri, Salar Bijari, Mohammad Reza Atashzar, Sajad P. Shayesteh, Bardia Khosravi, Mohammad Reza Babaei, Elnaz Jenabi, Mohammad Hasanian, Alireza Shahhamzeh, Seyaed Yaser Foroghi Ghomi, Abolfazl Mozafari, Arash Teimouri, Fatemeh Movaseghi, Azin Ahmari, Neda Goharpey, Rama Bozorgmehr, Hesamaddin Shirzad‐Aski, Roozbeh Mortazavi, Jalal Karimi, Nazanin Mortazavi, Sima Besharat, Mandana Afsharpad, Hamid Abdollahi, Parham Geramifar, Amir Reza Radmard, Hossein Arabi, Kiara Rezaei‐Kalantari, Mehrdad Oveisi, Arman Rahmim, Habib Zaidi
Abstract
BACKGROUND: We aimed to analyze the prognostic power of CT-based radiomics models using data of 14,339 COVID-19 patients. METHODS: 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. RESULTS: 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. CONCLUSION: 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.