VITALBEAT AI: PREDICTIVE CLASSIFICATION OF FETAL WELLBEING USING DATA ANALYTICS

Abstract

Fetal health monitoring is a vital component of prenatal care, as early detection of abnormal fetal conditions helps reduce risks to both mother and child. Cardiotocography (CTG) is commonly used to monitor Fetal Heart Rate (FHR) and uterine contractions; however, manual interpretation of CTG data requires expert knowledge and may lead to subjective judgment, inconsistencies, and delayed clinical decisions. These systems are prone to human error, limited scalability, inconsistent prediction outcomes, and difficulty handling imbalanced class distributions. Therefore, there is a strong need for an intelligent and automated system capable of providing accurate, fast, and reliable fetal health predictions. This research proposes a machine learning-based framework for automated classification of fetal cardiotocogram data into three categories: Normal, Suspect, and Pathological. The system incorporates preprocessing techniques such as missing value handling, categorical encoding, and class imbalance correction using the Synthetic Minority Over-sampling Technique (SMOTE). Multiple machine learning models, including Decision Tree Classifier (DTC), Logistic Regression Classifier (LRC), Gaussian Naive Bayes Classifier (GNBC), and a proposed Light Gradient Boosting Machine (LGBM), are implemented and evaluated to identify optimal performance. A Graphical User Interface (GUI) ensures accessibility and ease of use of research work. The proposed system enhances predictive accuracy, reduces reliance on subjective analysis, and provides consistent classification outcomes. By integrating advanced machine learning models with an interactive application interface, it supports clinical decision-making and contributes to intelligent healthcare solutions for effective antepartum monitoring.