Machine-Learning Approaches in COVID-19 Survival Analysis and Discharge-Time Likelihood Prediction Using Clinical Data

Abstract

As a highly contagious respiratory disease, COVID-19 has yielded high mortality rates since its emergence in December 2019. As the number of COVID-19 cases soars in epicenters, health officials are warning about the possibility of the designated treatment centers being overwhelmed by coronavirus patients. In this study, several computational techniques are implemented to analyze the survival characteristics of 1,182 patients. The computational results agree with the outcome reported in early clinical reports released for a group of patients from China that confirmed a higher mortality rate in men compared with women and in older age groups. The discharge-time prediction of COVID-19 patients was also evaluated using different machine-learning and statistical analysis methods. The results indicate that the Gradient Boosting survival model outperforms other models for patient survival prediction in this study. This research study is aimed to help health officials make more educated decisions during the outbreak.

Keywords: COVID-19; artificial intelligence; biostatistic; coronavirus; machine learning; pandemic; statistical analysis; survival analysis.

Graphical abstract

Figure 1

Probability Estimation of Discharge Time in Different Age and Sex Groups (A) Discharge-time probability estimation of sex groups after showing the symptoms. (B) Discharge-time probability estimation of sex groups after hospitalization. (C) Discharge-time probability estimation of two categories of age groups. (D) Discharge-time probability estimation of four categories of age groups.

Figure 2

Age Variation of 1,182 Patients Patients are categorized into four different age groups. First, second, and third quartiles are 34, 46, and 60, respectively.

Figure 3

Data-Processing Steps (A) Data collection and filtering. (B) Data-processing steps required for analysis.

Figure 4

Demonstration of Data Censorship Status Patients A, B, and D have not experienced any events until the end of the study, so they are considered as censored samples, but patient C is not censored because the event has occurred and it is fully observed.

Figure 5

Survival Analysis Algorithms Survival analysis techniques applied on COVID-19 data to predict survival time and hospital discharge-time probabilities.