Abstract

Background: Symptoms of Acute Respiratory infections (ARIs) among under-five children are a global health challenge. We aimed to train and evaluate the machine learning (ML) classification approaches in predicting symptoms of ARIs reported by mothers (caregivers) among children younger than five years in sub-Saharan African (sSA) countries.
Methods: We used the most recent (2012-2022) nationally representative Demographic and Health Surveys (DHS) data of 33 sSA countries. Machine learning algorithms such as logistic regression, LASSO, Elastic Net, Random Forest (RF), Artificial Neural Network (ANN), Naïve Bayes (NB), Boosting, and Bagged Tree were used for predicting the symptoms of ARIs among under-five children. We split the dataset into two datasets randomly, in which 80% of the data was used to train the model, and the remaining 20% was used to test the trained model. The model performance was evaluated using sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (AU-ROC).
Results: A total of 327,507 under-five children were included in the study. The prevalence of ARI was highest in Mozambique (15.4%), Uganda (15.2%), Togo (14.2%), and Namibia (13.8 %,), whereas Uganda (40.4%), Burundi (38.1%), Zimbabwe (37.4%), and Namibia (32.0%) had the highest prevalence of cough. The results of the random forest plot revealed that longitude, latitude, enhanced vegetation index, particulate matter, land surface temperature, nitrogen dioxide, and the number of cattle in the houses are the most important features in predicting the diagnosis of symptoms of ARIs among under-five children in sSA. The RF algorithm was selected as the best ML model (AUC=0.72, Accuracy=0.68) to predict the ARIs among children aged less than five.
Conclusions: Random forest machine learning algorithm (MLA) was classified as the best classifier to be employed for the prediction of the symptoms of ARI among under-five children.