A one-Dimensional Convolutional Neural Network Model for Predicting the Survival Outcome of Coronavirus Disease 2019

Coronavirus disease 2019 (COVID-19) is a pandemic that has spread rapidly and has been officially declared by the World Health Organization on 11 March 2020. However, there are few studies discussing the relationship between clinical procedures (i.e., tests, treatment, etc.) and the survival outcome of patients with this disease. In this study, we present a new 1D convolutional neural network (1D CNN) model for survival analysis and investigate the impact of clinical variables in predicting the survival outcomes of patients. We show that the proposed 1D CNN model is significantly better than the random forest model (AUC-ROC = 79.48% versus 72.48%) in classifying short-term from long-term survival groups of patients, with p