Fetal ECG extraction using multi-layer perceptron neural networks with Bayesian approach

In this paper, we investigate the application of neural networks to the problem of extracting fetal ECG from Maternal ECG early in pregnancy. The proposed extractor consists of a segmentation, feature extraction and Classification stage. For the feature extraction stage, coefficients of the Wavelet Transform (WT), real and imaginary parts of the Fast Fourier Transform and raw ECG data recorded from the abdomen of a pregnant woman were all found to be well-suited to FECG classification. Principal Component Analysis was used to reduce the dimensionality of the features. For the classification stage, Multi-Layer Perceptron neural networks were implemented according to Maximum Likelihood and Bayesian Learning formulations. The latter was found to make better use of training data and consequently produced better trained neural networks. Rejection thresholds of 0.9 were applied to the network output as a doubt level in order to ensure that only reliable classification decisions are made. A maximum classifier accuracy of 94.86% was achieved with 23.42% of patterns not being classified. Bayesian moderated outputs could not improve on these classification predictions significantly enough to warrant their added computational Overhead.