•  
  •  
 

Abstract

A fatal health condition involves an unborn baby that persists throughout the embryonic stage until delivery. The fetus grows and develops during each trimester of pregnancy. Obstetricians may detect fetal anomalies and select medical interventions based on cardiotocogram (CTG) data. However, the obstetrician's visual assessment of CTG data can sometimes be subjective or inaccurate. Therefore, automated analysis using machine learning approaches for CTG data is essential. This research employs decision analysis techniques, including decision trees (DT), gradient boosting (GB), and type-2 fuzzy neural networks (FNN), for prenatal analysis and prediction. The system was tested using a standard dataset consisting of three categories: normal health, fetal, and pathological. The dataset was divided into training and testing sets to evaluate fetal classification. In trials, the GB model achieved a high accuracy of 95%. Comparisons between the decision analysis methods and FNN models demonstrate the effectiveness of the decision analysis system. This technology can be integrated into clinical information systems to provide critical data on fetal health status.

Share

COinS