Fetal heart rate estimated from fetal electrocardiogram (fECG), recorded via electrodes placed on the maternal abdomen, represents an important tool in evaluating fetal distress and diagnosing fetal diseases. However, due to poor signal to noise ratio (SNR) of the fECG, the abdominal fECG analysis is still very difficult, despite of significant advances in technology and in the field of electrocardiography. In this study we propose a new efficient method for estimating the fetal QRS (fQRS) locations from abdominal signals (ADS). Thus, the method consists of several steps: pre-processing of the ADS in order to remove different types of noise including baseline wander and power-line interference (PLI); maternal electrocardiogram (mECG) cancellation using a simple template subtraction procedure and robust fQRS detections by employing empirical mode decomposition (EMD). This methodology was validated on real fECG recordings and its performance was compared with those obtained by two methods described in the literature. Four performance measures were defined: sensitivity (Se), positive predictive value (PPV), accuracy (ACC) and F1-measure. The proposed method overcomes the performance of the other two techniques described in literature.

Keywords: Fetal electrocardiogram, Empirical mode decomposition (EMD), Fetal QRS (fQRS) detection