Predictive Analytics for Improved Fetal Health Management

Fetal health monitoring is a critical aspect of prenatal care, ensuring the well-being of both mother and fetus. Traditional methods often involve invasive procedures, carrying risks and potential discomfort. In recent years, machine learning (ML) techniques have emerged as promising alternatives for non-invasive fetal health assessment. This study investigates the effectiveness of LightGBM, a gradient boosting framework, in predicting fetal health outcomes. By analyzing a dataset comprising various fetal and maternal parameters, this work compares LightGBM's performance against other popular ML algorithms, including CatBoost, XGBoost, SVM, Random Forest, Extra Trees, KNN, Logistic Regression, Naive Bayes, LDA, and QDA. Our findings demonstrate that LightGBM consistently outperforms these models in terms of accuracy, precision, recall, and F1-score, highlighting its potential as a valuable tool for early detection of fetal distress. This research contributes to the advancement of non-invasive fetal monitoring and provides insights for future applications in healthcare.