A Novel Triage Tool of Artificial Intelligence Assisted Diagnosis Aid System for Suspected COVID-19 Pneumonia in Fever Clinics
68 Pages Posted: 19 Mar 2020More...
Background: Currently, the prevention and control of COVID-19 outside Hubei province in China, and other countries has become more and more critically serious. We developed and validated a diagnosis aid model without CT images for early identification of suspected COVID-19 pneumonia (S-COVID-19-P) on admission in adult fever patients and made the validated model available via an online triage calculator.
Methods: Patients admitted from Jan 14 to Feb 26, 2020 with the epidemiological history of exposure to COVID-19 were included [Model development (n = 132) and validation (n = 32)]. Candidate features included clinical symptoms, routine laboratory tests and other clinical information on admission. Features selection and model development were based on Lasso regression. The primary outcome is the development and validation of a diagnosis aid model for S-COVID-19-P early identification on admission.
Findings: The development cohort contains 26 S-COVID-19-P and 7 confirmed COVID-19 pneumonia cases. The final selected features included 1 variables of demographic information, 4 variables of vital signs, 5 variables of blood routine values, 7 variables of clinical signs and symptoms, 1 infection-related biomarker. The model performance in held-out testing set and validation cohort resulted in AUCs of 0.841 and 0.938, F-1 score of 0.571 and 0.667, recall of 1.000 and 1.000, specificity of 0.727 and 0.778, and the precision of 0.400 and 0.500. The top 5 important features were Age, IL-6, SYS_BP, MONO%, and Fever classification. Based on this model, an optimized strategy for S-COVID-19-P early identification in fever clinics has also been designed.
Interpretation: S-COVID-19-P could be identified early by a machine-learning model only used collected clinical information without CT images on admission in fever clinics with 100% recall score. The well performed and validated model has been deployed as an online triage tool, which is available at: https://intensivecare.shinyapps.io/COVID19/ .
Funding Statement: The present study was supported by grants from the PLA Science and Technology Project (14CXZ005, AWS15J004, 16BJZ19), National Key R&D Program of China (2019YFF0302300), Construction Project of Key Disciplines in the 13th Five-Year Plan of the PLA (Traumatic Surgery in the Battlefield, 2019-126, 2019-513), Beijing Science and Technology New Star Project (XX2018019/Z181100006218028), the PLA General Hospital Science and technology Project (2019XXJSYX20, 2018XXFC-20, ZH19016).
Declaration of Interests: The authors declare that they have no conflict of interest.
Ethics Approval Statement: Data collection was passive and had no impact on patient safety. This study was approved by the PLA General Hospital ethics committee.
Keywords: Suspected COVID-19 pneumonia; Diagnosis Aid model; Fever Clinics; Machine Learning
Suggested Citation: Suggested Citation