An artificial neural network model to predict the mortality of COVID-19 patients using routine blood samples at the time of hospital admission

Abstract

Background: In a pandemic situation (e.g., COVID-19), the most important issue is to select patients at risk of high mortality at an early stage and to provide appropriate treatments. However, a few studies applied the model to predict in-hospital mortality using routine blood samples at the time of hospital admission. This study aimed to develop an app, name predict the mortality of COVID-19 patients (PMCP) app, to predict the mortality of COVID-19 patients at hospital-admission time. Methods: We downloaded patient records from 2 studies, including 361 COVID-19 patients in Wuhan, China, and 106 COVID-19 patients in 3 Korean medical institutions. A total of 30 feature variables were retrieved, consisting of 28 blood biomarkers and 2 demographic variables (i.e., age and gender) of patients. Two models, namely, artificial neural network (ANN) and convolutional neural network (CNN), were compared with each other across 2 scenarios using raw laboratory versus normalized data and training vs testing datasets (n = 361 and n = 106/361≅30%) to verify the model performance (e.g., sensitivity [SENS], specificity [SPEC], and area under the receiver operating characteristic curve [AUC]). An app for predicting the mortality of COVID-19 patients was developed using the model's estimated parameters for the prediction and classification of PMCP at an earlier stage. Feature variables and prediction results were visualized using the forest plot and category probability curves shown on Google Maps. Results: We observed that the normalized dataset gains a relatively higher AUC(>0.9) when compared to that(<0.9) in the raw-laboratory dataset based on training data, the normalized dataset in ANN yielded a high AUC of 0.96 that that(=0.91) in CNN based on testing data, and a ready and available app, where anyone can access the model to predict mortality, for PMCP was developed in this study. Conclusions: Our new PMCP app with ANN model accurately predicts the mortality probability for COVID-19 patients. It is publicly available and aims to help health care providers fight COVID-19 and improve patients’ classifications against treatment risk.

Design of an artificial neural network to predict mortality among COVID-19 patients

“…Asteris et al (2022) trained four ML techniques on the data of 10,237 patients, and finally implemented and evaluated the ANN model to predict mortality in COVID-19 patients with an accuracy of 89.47% [ 75 ]. The ANN model developed by Lin et al (2021) predicts the mortality risk of COVID-19 patients with an AUC of 0.96 [ 76 ]. Adib et al (2021) also compared three ML models' performance for mortality analysis of pregnant women with COVID-19.…”

Mortality prediction of COVID-19 patients using soft voting classifier

“…A Dataset consisting of 114 patients who tested positive for COVID was trained and evaluated using 5-fold K-fold cross-validation on a hyperparameter-tuned XGBoost model ( Ryan et al, 2020 ). A combination of data from 361 patients from Wuhan, China and 106 patients from Korean medical institutions was utilized to train and test an artificial neural network in paper ( Lin, Chien, Wang, and Chou, 2021 ). With data obtained from five hospitals in New York City for 4098 Covid-19 patients, Extreme Gradient Boosting (XGBoost) was utilized to predict in-hospital mortality at time intervals of 3,5,7, and 10 days from admission ( Vaid et al, 2020 ).…”

“…On normalized data, the ANN had an accuracy of 80% and the CNN had an accuracy of 73%. Age, gender, and the number of lymphocytes and neutrophils present in the blood were found to be the most important factors to determine the mortality of the patient ( Lin, Chien, Wang, and Chou, 2021 ). A study has been conducted utilising clinical data of patients on 73 blood biomarkers.…”

Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran

“…Their model AUC was 0.95 [ 46 ]. In a study by Lin et al of 30 demographic and laboratory biomarkers using machine learning methods including neural network to predict mortality of COVID-19 patients, the accuracy and AUC of this model were 0.91 and 0.88, respectively [ 47 ]. In the study of Morales et al, ten important demographic and laboratory biomarkers were used, including age, blood pressure, liver, and kidney failure.…”