Purpose: to develop a tool for early diagnosis of SARS‐CoV2‐infected patientsDesign: cross‐sectional, retrospective, single‐centre (2 time frame study: training ‐ validation data set)Recruitment: 1311 patients who presented to the First Affiliated Hospital, School of Medicine, Zhejiang University with at least 1 SARS‐CoV‐2 RT‐PCR testSample size: n = 304 (73 cases) (= subset of the study including training dataset only)n = 95 (18 cases) (= validation dataset)Inclusion criteriaAll RT‐PCR‐positive cases; 1311
All RT‐PCR‐negative patients who came to the First Affiliated Hospital, School of Medicine, Zhejiang University and performed with at least 1 SARS‐CoV‐2 nucleic acid detection for analysis RT‐PCR
First 60% of negative outpatients sorted by 'Z‐A' based on Chinese first name from Qingchun District (training dataset), and then final 40% who presented (validation dataset)
Exclusion criteriaAsymptomatic patients without history of exposure but had strong willingness for detection
Patients with "important" missing data