Comparative Analysis of Machine Learning Algorithms for Enhanced Clinical Prediction of Fetal Health Status

The Child and Maternal deaths during pregnancy are one of the major catastrophes occurring globally. Reduction of child mortality is a tragic issue in several of the United Nations Sustainable Development Goals and is a key indicator of human progress. This research aims to assess the efficacy of Machine Learning algorithms, including K-Nearest Neighbor (KNN), Support Vector Machines (SVM’s), Random Forest (RF) and Gradient Boosting (GB) in accurately predicting fetal health status by making use of Cardiotocograms (CTGs) technology. By comparing these algorithms, we seek to enhance CTG technology's ability to monitor the clinical parameters of a fetus such as baseline value, fetal movements, uterine contractions and reduce the risk of child and maternal mortality. This study underscores the importance of leveraging advanced computational methods to bridge gaps in healthcare thereby addressing critical healthcare challenges and ultimately striving towards more equitable maternal and child health outcomes globally. The proposed research has been tested with 2113 records obtained from Kaggle and has been successful in predicting the best algorithm with an accuracy of 96.926%.