Robust and Reliable Assessment of Preeclampsia Using Statistical Machine Learning

Artificial intelligence is increasingly prevalent, in today's world. Machine learning techniques play a role, in the healthcare field as a whole and particularly, in the areas of obstetrics and gynecology. More specifically, methods in machine learning can be used to enhance the health and well-being of pregnant women, closely monitoring their health parameters during pregnancy. In this paper, we present a solution for predicting the risk of preeclampsia, a life-threatening disorder affecting women during pregnancy. For this purpose, we collected data from three gynaecologist in Pakistan providing health services in low resource settings in which data has samples of 233 pregnant women using 13 features which are analysed during routine checkups. We adopted model comparison approach where we trained and tested our data on multiple classification models. Performance of each model was tested in terms of accuracy, precision, recall and F1-score. On the basis of these metrics, logistic regression and XGBoost were found to be the best performing models with the accuracy of 85%.