Machine learning application for the prediction of SARS-CoV-2 infection using blood tests and chest radiograph

Du, Richard; Tsougenis, Efstratios D.; Ho, Joshua W. K.; Chan, Joyce K. Y.; Chiu, Keith W. H.; Fang, Benjamin X. H.; Ng, Ming Yen; Leung, Siu-Ting; Lo, Christine S. Y.; Wong, Ho-Yuen F.; Lam, Hiu-Yin S.; Chiu, Long-Fung J.; So, Tiffany Y; Wong, Ka Tak; Wong, Yiu Chung I.; Yu, Kevin; Yeung, Yiu-Cheong; Chik, Thomas; Pang, Joanna W. K.; Wai, Abraham Ka-chung; Kuo, Michael D.; Lam, Tina P. W.; Khong, Pek-Lan; Cheung, Ngai-Tseung; Vardhanabhuti, Varut

Triaging and prioritising patients for RT-PCR test had been essential in the management of COVID-19 in resource-scarce countries. In this study, we applied machine learning (ML) to the task of detection of SARS-CoV-2 infection using basic laboratory markers. We performed the statistical analysis and trained an ML model on a retrospective cohort of 5148 patients from 24 hospitals in Hong Kong to classify COVID-19 and other aetiology of pneumonia. We validated the model on three temporal validation sets from different waves of infection in Hong Kong. For predicting SARS-CoV-2 infection, the ML model achieved high AUCs and specificity but low sensitivity in all three validation sets (AUC: 89.9–95.8%; Sensitivity: 55.5–77.8%; Specificity: 91.5–98.3%). When used in adjunction with radiologist interpretations of chest radiographs, the sensitivity was over 90% while keeping moderate specificity. Our study showed that machine learning model based on readily available laboratory markers could achieve high accuracy in predicting SARS-CoV-2 infection.