Fetal Health Classification and Birth Weight Estimation Using Machine Learning

Abstract

This paper addresses fetal health prediction via Cardiotocography (CTG) data analysis, utilizing models like Logistic Regression, SVM, and boosting algorithms. Feature selection methods such as PCA and LDA are employed, along with SMOTE for dataset balancing. CatBoost model emerges as superior with 99% accuracy. Fetal weight prediction remains challenging, tackled through machine learning algorithms incorporating parameters like gestational period and maternal factors. Models like Random Forest and Adaboost are employed, with RMSE analysis guiding their combination for improved prediction accuracy. Overall, the paper emphasizes leveraging ML for fetal health classification and birth weight prediction.