The maternal and fetal health dilemmas at the international level are henceforth highlighted by the statistics that revolve around a shocking 6.7 million fetal intrauterine deaths routinely. The research shed light on the critical importance of NST (nonstress test) to evaluate and aid in the early detection of fetal problems is a centerpiece of this research. This study introduces an innovative predictive method based on ensemble learning, aiming to classify fetal health into three categories: normal, suspicious, and pathology. Making use of a large cardiotocography dataset that includes heart rate acceleration and fetal movement from NST tests, our approach combines sophisticated analytics with clinical data. As a part of ensemble learning, decision trees stand out as flawless flowcharts with an exemplary capability to filter and classify fetal health data for potential risks and improved health outcomes. The ensemble of their deeds together creates a layer of sophistication, allowing for several decision trees to cooperate and make a well-informed and accurate classification. The integration of XGBoost (Extreme Gradient Boosting), further enhances predictive prowess by optimizing decision tree strengths, refining their abilities, and minimizing errors. The study of relationship between the maternal hormones and by offering a unique method, the fetal development aims to enhance early diagnosis and classification of fetal health status. The collaborative power of decision trees and XGBoost within the ensemble not only advances the field of machine learning in healthcare but also has the potential to save lives by enabling proactive maternal and fetal healthcare measures.