Foetal Health Prediction using Random Forest and Support Vector Machine

Abstract

In order to identify any potential hazards to the foetus and take the necessary precautions to guarantee a healthy birth, foetal health prediction is a crucial component of maternal healthcare. In this paper, we evaluate the application of Random Forest and Support Vector Machine as machine learning algorithms for predicting foetal health based on various foetal health data. 2,114 records from cardiotocography (CTG) assessments of foetal heart rate and uterine contractions make up the dataset used in this investigation. Using a variety of performance indicators, including accuracy, precision, recall, and F1-score, we assess the abilities of Random Forest and Support Vector Machine to predict foetal health. In order to diagnose and treat foetuses who may have health issues early on, it is essential to do research into foetal health prediction. In this study, we examine the performance of Support Vector Machine and Random Forest, two well-known machine learning algorithms, at predicting foetal health. To train and evaluate our models, we used the CTG dataset, which includes foetal heart rate recordings. Our findings show that both algorithms are highly accurate at predicting foetal health, with Random Forest slightly outperforming Support Vector Machine in this regard. Furthermore, our research demonstrates that in terms of precision, recall, and F1-score, the Random Forest approach surpasses the Support Vector Machine technique. Based on our research, Random Forest might be a good algorithm for predicting foetal health in clinical situations.