Evaluation of fetal heart rate baseline estimation method using testing signals based on a statistical model.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Medical Informatics

Background:

• Computer-aided fetal monitoring relies on automated analysis of fetal heart rate (FHR) variability.
• Accurate estimation of the FHR baseline is a critical first step in automated FHR signal interpretation.
• Existing baseline estimation algorithms vary in efficiency.

Purpose of the Study:

• To evaluate and compare the efficiency of different FHR baseline estimation algorithms.
• To assess the performance of an artificial neural network (ANN) based algorithm against a classical nonlinear filtering approach.
• To validate algorithm performance using a signal modeling method with a preset baseline component.

Main Methods:

• Developed a method for modeling FHR signals based on a preset baseline component for algorithm evaluation.
• Generated synthetic FHR signals using known baseline components.
• Compared baseline estimations from an ANN algorithm and a classical nonlinear filtering algorithm against the known component baselines.

Main Results:

• The ANN-based algorithm demonstrated superior performance in estimating the FHR baseline.
• The ANN algorithm's estimated baselines closely matched the preset component baselines used in signal modeling.
• The classical nonlinear filtering algorithm showed lower accuracy in baseline estimation.

Conclusions:

• Artificial neural networks offer a more effective approach for FHR baseline estimation in computer-aided fetal monitoring.
• The developed signal modeling method provides a robust framework for evaluating FHR baseline estimation algorithms.
• Improved FHR baseline estimation can enhance the accuracy of automated fetal monitoring systems.