International Research Journal of Education and Technology

Background: Each year, approximately 6.3 million fetal deaths occur during pregnancy, according to the World Health Organization (WHO). A nonstress test (NST) is one of the most widely used methods to monitor and diagnose potential risks to fetal and maternal health. It analyzes fetal heart rate and uterine contractions, with results typically interpreted by experts from printed NST traces. (2) Methods: This study introduces a machine learning-based approach, utilizing ensemble learning to classify fetal health (normal, suspicious, or pathological) using a dataset containing fetal heart rate and movement data from NST tests. (3) Results: The developed model achieved an accuracy of over 99.5% on the test data. (4) Conclusions: Experimental findings demonstrate that the machine learning model can effectively be used to assist in diagnosing fetal health during NST evaluations.

Keywords: ensemble learning, fetal health, fetal heart rate, NS