Cardiotocography Data Analysis to Predict Fetal Health Risks with Tree-Based Ensemble Learning

A sizeable number of women face difficulties during pregnancy, which eventually can lead the fetus towards serious health problems. However, early detection these risks save both invaluable life infants and mothers. Cardiotocography (CTG) data provides sophisticated information by monitoring heart rate signal fetus, is used to predict potential fetal wellbeing for making clinical conclusions. This paper proposed analyze antepartum CTG (available on UCI Machine Learning Repository) develop an efficient tree-based ensemble learning (EL) classifier model status. In this study, EL considers Stacking approach, a concise overview approach discussed developed accordingly. The study also endeavors apply distinct machine algorithmic techniques dataset determine their performances. technique, in paper, involves four algorithms, namely, Random Forest classifier, Decision Tree Extra Trees Deep as base learners. contains 21 features, but only 10 most important features are selected from with Chi-square method experiment, then normalized Min-Max scaling. Following that, Grid Search applied tuning hyperparameters algorithms. Subsequently, 10-folds cross validation performed select meta learner model. comparative assessment made between individual algorithms model; finding depicts classifiers’ superiority prediction securing accuracy about 96.05%. Eventually, concludes that be substantial paradigm studies improve models’ reduce error rate.