Wavelet-Based Post-Processing Methods for the Enhancement of Non-Invasive Fetal ECG

Abstract

Despite the number of techniques developed in the literature, the extraction of a clean fetal ECG (fECG) from non-invasive recordings is still an open research issue. In this work, different wavelet-based post-processing approaches for the denoising of the fECG were evaluated. A small dataset composed of twenty signals recorded from ten pregnant women between the 21st and the 27th week of gestation was adopted. fECG extraction was accomplished by using a multireference QR-decomposition-based recursive least squares adaptive filter. Then, all signals were decomposed with the stationary wavelet transform (SWT) and stationary wavelet packet transform (SWPT), using a 7-level decomposition with Haar mother wavelet and hard-thresholding. Two different thresholds from the literature were tested: the first one is level-independent (Minimax) while the other one is level-dependent. The latter was adapted to be exploited on SWPT. The enhancement of the fetal QRS complex was analyzed by computing the improvement of the signal-to-noise ratio and the performance of a fetal QRS detector. The comparative analysis revealed how the SWT outperforms the more complex SWPT, regardless the thresholding approach.