Rapid and accurate identification of COVID-19 infection through machine learning based on clinical available blood test results

Jiangpeng Wu; Pengyi Zhang; Liting Zhang; Wenbo Meng; Junfeng Li; Chongxiang Tong; Yonghong Li; Jing Cai; Zengwei Yang; Jinhong Zhu; Meie Zhao; Huirong Huang; Xiaodong Xie; Shuyan Li

Currently, suspected patients were recommended to use nucleic acid assay to effectively control the spread of the epidemic pneumonia. The complexity of the operation and the fragility of the specimens make it unsuitable for large-scale initial screening. The time-consuming etiology testing might delay the timely isolation and treatment for COVID-19 patients. And a study reported that the positive rate of the testing was only 30-50% in clinical application. 9, 28 The sequencing technology of the viral had elevated the accuracy of detection results, but it needed high-quality laboratory requirement which could only be provided for supplementary evidence. Rapid antibody detection often lagged behind the nucleic acid detection due to the presence of the window period. 14 After the virus invaded, the dynamic . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.