Model E9

Model E9

Description of the tool
Diagnostic nomogram from China to predict COVID-19 pneumonia using semantic features from chest CT. DOI: https://doi.org/10.1101/2020.04.03.20052068

To which patients does this tool apply?
This tool was developed based on patients suspected of COVID-19 pneumonia. Suspected diagnostic criteria are:1) epidemiological history, 2) fever and / or respiratory symptoms, 3) suspected COVID-19 pneumonia imaging features, 4) normal range or decrease of white blood cell (WBC) count or lymphocyte at beginning of disease. Patients who have epidemiological history + any two clinical symptoms or no epidemiological history + three clinical symptoms can be suspected of COVID-19 pneumonia
no      yes
no      yes
no      yes
no      yes
no      yes
≥4e9/L      <4e9/L

What information will you need?

Epidemiological history, semantic features from chest CT (wedge-shaped or fan-shaped lesion parallel to or near the pleura, bilateral lower lobes, ground glass opacities, crazy paving pattern), and white blood cell (WBC) count.

How was this tool developed?

The primary cohort (Wenzhou, China) consisted of 178 suspected COVID-19 pneumonia patients, 89 of whom were positive. The validation cohort (Yueqing, China) consisted of 116 suspected COVID-19 pneumonia patients, 68 of them positive. All patients with suspected COVID-19 pneumonia had throat swab samples or sputum sample taken at admission (at least two samples were taken, at least 24 hours apart) to establish if they were positive. LASSO was used for feature selection; features were combined into a multivariable logistic regression model. The C-statistic of the nomogram was 0.848 in external validation cohort.