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

Computer-aided fetal monitoring is based on automated analysis of the fetal heart rate (FUR) variability. The first and the main step in the automated signal interpretation is the estimation of the so called FHR baseline. There are various algorithms for baseline estimation, of different efficiency. For its evaluation, the method of modeling of FHR signal based on the preset baseline component has been developed. The best algorithm is expected to provide the same baseline as the component baseline used to model the FHR signal. Generated signals were used to compare the baselines that have been estimated by two algorithm: the first one relying on artificial neural networks and the classical one using nonlinear filtering of FHR signal.