Automatic diagnosis of fetal heart rate: comparison of different methodological approaches

The cardiotocograpy (CTG) is the clinical, traditional, noninvasive approach to monitor the fetal condition antepartum. CTG analysis is focused on the detection of fetal heart rate parameters from which the clinicians can identify by eye inspection some patterns associated to fetal activity. However this qualitative method rarely can detect the emergence of fetal pathologies. This study aims at finding new algorithms which can enhance the differences among the normal CTG signals and those presenting anomalies due to a pathological status. On a database of more than 500 recordings, we tested different classification methods to identify normals from potential pathological fetuses. A Multilayer Perceptron (MLP) neural network and an Adaptive Neuro-Fuzzy Inference System (ANFIS) were compared with classical statistical methods. Both the neural and neuro-fuzzy approaches seem to give better results than any tested statistical classifier.