Predicting Maternal Health Risks Using Machine Learning: A Comparative Study of Classification Algorithms

Abstract

Maternal health is a critical aspect of public healthcare, with complications during pregnancy and childbirth contributing to high mortality rates worldwide. This study explores the potential of machine learning techniques in predicting maternal health risks using physiological indicators such as age, blood pressure, blood sugar levels, body temperature, and heart rate. A dataset from Kaggle was preprocessed to ensure quality, followed by exploratory data analysis to identify key patterns. Three machine learning models: Random Forest, Support Vector Machines, and K-Nearest Neighbors were trained and evaluated. The Random Forest classifier emerged as the best-performing model, achieving an accuracy of 80.79%, along with superior precision, recall, and F1-score. Performance evaluation using confusion matrices confirmed its effectiveness in distinguishing maternal risk levels. These findings highlight the potential of machine learning for early risk assessment, supporting timely medical interventions. While the study demonstrates promising results, further improvements such as model optimization, dataset expansion, and integration into real-world healthcare systems are necessary to maximize impact. This research contributes to the growing field of AI-driven maternal healthcare, paving the way for predictive tools that could enhance clinical decision-making and improve maternal health outcomes.