FetalCare: Enhancing Prenatal Care through a Machine Learning Approach for Fetal Health Classification and Birth Weight Prediction

The process of navigating the intricate landscape of maternal and infant health, allows the research work described in this paper to confront the formidable challenge of preterm birth risk through a sophisticated machine learning framework tailored for prenatal care. The focus on both fetal health classification and birth weight prediction allows this work to employ different algorithms—Support Vector Classifier, Random Forest, AdaBoost, and Decision Tree by seamlessly integrating them through a refined ensemble approach. This strategic fusion prioritizes precision in categorizing fetal health into “NORMAL,” “SUSPECT,” or “PATHOLOGICAL,” reflecting an unwavering commitment to accurate identification, especially in critical health conditions. Rigorous evaluation metrics, notably the macro recall score, underscore the dedication to robust assessments essential in addressing the multifaceted nature of preterm birth risks. Simultaneously, the birth weight prediction component employs a combination of AdaBoost and Random Forest regressors, further enhancing the accuracy of estimations.