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

The most common laboratory findings, including hematological and biochemical parameters, play a prominent part in the initial screening for COVID-19. 22 In the present work, we use machine learning algorithm to process these valuable parameters in order to realize the progress of accurate and rapid identification for the emerging pneumonia. The proposed assistant discrimination tool is more convenient and practical than nucleic acid detection and antibody detection, which can accelerate the recognition of suspected patients and thus reduce the risk of transmission. Moreover, the deep relationship between these important reference targets and COVID-19 could help to monitor the progress of the disease and to guide useful strategies for clinicians. Finally, we aim to design a simple and effective pre-warning system to help proactively initial screening for suspected patients which will contribute to the outbreak control and public health.