Fetal Phonocardiogram Decomposition Framework

In this paper, an efficient procedure to identify vital bio-signals from the Fetal Phonocardiogram (FPCG) signal is presented. FPCG signal is rich in physiological bio-signals. However, extracting these bio-signals with fidelity is a challenging task since they overlap in time and frequency and as well as corrupted by noise induced from the maternal and fetal movements. While current approaches use a two-step process of de-noising the signal and then extract the vital bio-signals, Empirical Mode Decomposition (EMD) algorithm can efficiently de-noise and extract the maternal and fetal heart sounds in a single step. Frequency range obtained from the Short Time Fourier Transform (STFT) is used to validate the algorithm. It is observed that the bandwidths of extracted FHS and MHS signals are consistent with the existing gold standards.