Evaluation of cyclic repetition frequency based algorithm for fetal heart rate extraction from fetal phonocardiography

The passive and non-invasive approach of fetal heart rate (FHR) estimation using fetal phonocardiography (fPCG) provides an alternative to the commonly used Doppler ultrasound-based technique. The acoustic signal recorded on the mother's abdomen is corrupted by noise, making FHR estimation difficult. Thus, an accurate and reliable FHR measurement algorithm from the fPCG signal is needed. This study evaluates a cyclic repetition frequency-based algorithm with additional FHR validation steps on the clinical recordings from three datasets with varying signal quality. The FHR from the algorithms is compared against the reference FHR measurements, and the accuracy and reliability performance measure is determined. The cyclic repetition algorithm significantly improves (p < 0.05) against the Hilbert-envelope autocorrelation algorithm for poor-quality signals. The algorithm's average reliability across three datasets is 93%, and accuracy is 4.5 beats per minute. The algorithm performs well for the gestation week ranging from 26-40 and for the labor and antenatal ward data.