Fetal Health Status Classification Using Machine Learning Algorithms-A Comparative Analysis

Fetal health status classification is essential for monitoring and confirming the well-being of the fetus through pregnancy. This study evaluates various Machine Learning techniques for classifying fetal health status using the Cardiotocography (CTG) dataset from the University of California Irvine Machine Learning Repository(UCI) Repository and the same is available in in Kaggle community also. The dataset is obtained by an automated processing system which was used to measure the diagnostic features of 2126 fetal Cardiotocograms (CTGs) and a consensus classification label(normal, risky and pathological) was issued to each of the CTGs by three expert obstetricians in this dataset. Assessment and comparison of the performance of multiple models, including Decision Tree (DT), Gradient Boosting (GB), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Stochastic Gradient Descent (SGD), AdaBoost, and XGBoost are performed. Key performance indicators like accuracy, precision, recall, and F-measure are calculated for each model to determine their effectiveness in fetal health classification. The results reveal significant variations in performance across the models, with certain techniques demonstrating superior classification accuracy and robustness. This comparative analysis provides insights into the strengths and limitations of several Machine Learning (ML) approaches for fetal health status classification, guiding future research and practical applications in prenatal care.