Active Learning Technique for Fetal Health Classification

Fetal health classification is a crucial aspect of prenatal care, enabling early detection of potential complications. This project presents an active learning-based approach for classifying fetal health status using machine learning algorithms. The dataset used in this research includes various features such as fetal heart rate and other physiological parameters to predict fetal health outcomes. Several models, including XGBoost, CatBoost, Random Forest, Stacking Ensemble, MLP, and SVM, are employed for the classification task. The application integrates SHAP (SHapley Additive exPlanations) plots for model interpretation, helping to explain the decision-making process behind each classification. The front-end of the system is built using HTML, CSS, and JavaScript, while the back-end leverages Python with Flask. This research aims to enhance the accuracy of fetal health predictions and improve the interpretability of machine learning models used in medical decision-making.

Keywords: Fetal Health, Machine Learning, Active Learning, Classification, SHAP, XGBoost, CatBoost, Random Forest, Predictive Modeling, Flask