Risk Assessment of Pregnancy-induced Hypertension Using a Machine Learning Approach

Abstract

This research aimed to develop a predictive model of the risk assessment of pregnancy-induced hypertension using a machine learning approach. Pregnancy-induced hypertension is a complication that has a serious impact on pregnant women and fetuses. It is the world's top three cause of death among pregnant women [1]. Nowadays, the exact cause of pregnancyinduced hypertension is unknown and therefore cannot be prevented. Early detection and received treatment can reduce the severity and danger. A public dataset of Logan (2020) was used in this research [2] the dataset was collected from a casecontrol study on the determinants of 83 pre-eclampsia and five eclampsia cases among 352 pregnant women delivering in county hospitals in Nairobi, Kenya. According to the dataset, 75 percent of the pregnant women were healthy. Only 25 percent of the pregnant women were pre-eclampsia and eclampsia. Thus, this would result in a problem of an imbalanced classification when one of the two classes had more data than the other class. As such, this problem was resolved with the Synthetic Minority Over-sampling Technique (SMOTE). Risk assessment of pregnancy-induced hypertension was performed on seven machine learning algorithms, which were logistic regression (LR), K-nearest neighbor (KNN), decision tree (DT), random forest (RF), multilayer perceptron neural network(MLP), support vector machines (SVM), and naive Bayes (NB). In the experimental results, RF had the highest accuracy at 89.62 percent compared to other machine learning algorithms.