Abstract
According to the World Health Organization (WHO), a significant reduction in mortality and maternal morbidity has occurred in developed countries over the past decades. In contrast, these rates remain high in developing countries. Smart mobile-health (m-health) applications that use machine learning (ML) approaches are necessary tools for pregnancy monitoring in an accessible, reliable, and cost-efficient manner, making the prediction of high-risk situations possible during gestation. This paper, therefore, proposes the development, performance evaluation, and comparison of ML algorithms based on Bayesian networks capable of identifying at-risk pregnancies based on the symptoms and risk factors presented by the patients. A performance comparison of several Bayes-based ML algorithms determined the best-suited algorithm for the prediction, identification, and accompaniment of hypertensive disorders during pregnancy. The contribution of this study focuses on finding a smart classifier for the development of novel mobile devices, which presents reliable results in the identification of problems related to pregnancy. Through the well-known cross-validation method, this proposal is evaluated and compared with other recent approaches. The averaged one-dependence estimators presented better results on average than the other approaches. These findings are key to improving the health monitoring of women suffering from high-risk pregnancies around the world. Thus, this study can contribute to a reduction of both maternal and fetal deaths.