<![CDATA[This study aims to evaluate and compare the performance of Linear Regression and Random Forest Regression in predicting fetal health conditions using cardiotocography (CTG) data obtained from the Kaggle Fetal Health dataset. The dataset contains 2,126 CTG records with 22 physiological attributes, including baseline fetal heart rate, accelerations, fetal movements, uterine contractions, short-term variability, long-term variability, and histogram-based measures. These features represent key indicators used in clinical fetal monitoring and provide a comprehensive basis for quantitative prediction. The research methodology includes data preprocessing, normalization, training–testing partition, model development, and performance evaluation using Mean Absolute Error, Mean Squared Error, Root Mean Squared Error, and the coefficient of determination. The results show that Random Forest Regression achieves significantly better performance than Linear Regression, offering lower error values and higher predictive accuracy. This is likely due to its capability to model nonlinear relationships and complex interactions inherent in CTG data. In contrast, Linear Regression demonstrates limited capability in capturing these patterns, resulting in reduced predictive quality. Overall, the findings indicate that ensemble-based approaches such as Random Forest are more suitable for fetal health prediction and can support the development of reliable decision-support tools in obstetric care. The study contributes to the advancement of data-driven fetal monitoring by presenting a comparative analysis of regression methods for predicting fetal health scores.]]>
