Wavelet-based denoising of fetal phonocardiographic signals

摘要

Auscultation is still one of the first basic analytical tools used to evaluate functional state of the fetal heart, as well as the first indicator of fetal well-being. Its modern form is called fetal phonocardiography (fPCG). The fPCG technique is passive and can be used for long-term monitoring. In order to improve the diagnostic capabilities of fPCG, robust signal processing techniques are needed for denoising of the signals. Traditional denoising techniques apply a linear filter to remove the noise and interference from the fPCG signals. These methods have certain limitations for the non-stationary random fPCG signals. In this paper, an improved technique for denoising of fPCG signals is presented. A highly sensitive data recording module is used to acquire the fPCG signals from the maternal abdominal surface. The acquired fPCG signals are decomposed, denoised and reconstructed by utilising Matlab wavelet transform toolbox. The proposed approach improves the signal to noise ratio (SNR) of these signals. The presented technique can be used in preprocessing stage of all fPCG-based fetal monitoring applications.