Computational Learning Approaches for Personalized Pregnancy Care

Abstract

The increasing use of interconnected sensors to monitor patients with chronic diseases, integrated with tools for the management of shared information, can guarantee a better performance of health information systems (HISs) by performing personalized healthcare. The early diagnosis of chronic diseases such as hypertensive disorders of pregnancy represents a significant challenge in women’s healthcare. Computational learning techniques are useful tools for pattern recognition in the assessment of an increasing amount of integrated data related to these diseases. Hence, in this paper, the use of machine learning (ML) techniques is proposed for the assessment of real data referred to hypertensive disorders in pregnancy. The results show that the averaged one-dependence estimator algorithm can help in the decision- making process in uncertain moments, thus improving the early detection of these chronic diseases. The best-evaluated computational learning algorithm improves the performance of HISs through its precise diagnosis. This method can be applied in electronic health (e-health) environments as a useful tool for handling uncertainty in the decision-making process related to high-risk pregnancy.