Radiomics‐based model for accurately distinguishing between severe acute respiratory syndrome associated coronavirus 2 (SARS‐CoV‐2) and influenza A infected pneumonia

• DOI: 10.1002/mco2.14
• Publication Date: 2020-08-13T13:11:36Z
• Authors: Qi‐Qiang Zeng; Kenneth I. Zheng; Jun Chen; Zheng‐Hao Jiang; Tian Tian; Xiao‐Bo Wang; Hong‐Lei Ma; Ke‐Hua Pan; Yun‐Jun Yang; Yong‐Ping Chen; Ming‐Hua Zheng
• Keywords: Nomogram; Viral Pneumonia

Clinicians have been faced with the challenge of differentiating between severe acute respiratory syndrome associated coronavirus 2 (SARS-CoV-2) infected pneumonia (NCP) and influenza A (IAP), a seasonal disease that coincided outbreak. We aim to develop machine-learning algorithm based on radiomics distinguish NCP from IAP by texture analysis computed tomography (CT) imaging. Forty-one 37 patients admitted January February 6, 2019 two hospitals in Wenzhou, China. All had undergone chest CT examination blood routine tests prior receiving medical treatment. was diagnosed real-time RT-PCR assays. Eight 56 radiomic features extracted LIFEx were selected least absolute shrinkage selection operator regression score subsequently constructed into nomogram predict area under operating characteristics curve 0.87 (95% confidence interval: 0.77-0.93). The also showed excellent calibration Hosmer-Lemeshow test yielding nonsignificant statistic (P = .904). novel may efficiently patients. be incorporated existing diagnostic effectively stratify suspected for SARS-CoV-2 pneumonia.