Combination of four clinical indicators predicts the severe/critical symptom of patients infected COVID-19.

摘要: Abstract Background Despite the death rate of COVID-19 is less than 3%, fatality severe/critical cases high, according to World Health Organization (WHO). Thus, screening before symptom occurs effectively saves medical resources. Methods and materials In this study, all 336 patients infected in Shanghai March 12th, were retrospectively enrolled, divided training test datasets. addition, 220 clinical laboratory observations/records also collected. Clinical indicators associated with symptoms identified a model for prediction was developed. Results Totally, 36 significantly identified. The are mainly thyroxine, immune related cells products. Support Vector Machine (SVM) optimized combination age, GSH, CD3 ratio total protein has good performance discriminating mild cases. area under receiving operating curve (AUROC) reached 0.9996 0.9757 testing dataset, respectively. When using cut-off value as 0.0667, recall 93.33 % 100 datasets, separately. Cox multivariate regression survival analyses revealed that discriminated used information selected indicators. Conclusion robust effective predicting COVID