Non-invasive FECG extraction from a set of abdominal sensors

Introduction: Despite significant advances in adult clinical electrocardiography (ECG) signal processing techniques and the power of digital processors, the analysis of non-invasive foetal ECG (NI-FECG) is still in its infancy. The PhysioNet/Computing in Cardiology Challenge 2013 addresses some of these limitations by making publicly available a set of FECG data to the scientific community for evaluation of signal processing techniques. Methods: ECGs were first preprocessed by cascading a low pass and a high pass filter in order to remove higher frequency and baseline wander. A Notch filter to remove power interferences at 50Hz or 60Hz was applied if required. The signals were then normalised before applying various source separation techniques to cancel the maternal ECG. These techniques included: template subtraction, principal/independent component analysis, extended Kalman filter and a combination of a subset of these methods (FUSE method). FQRS detection was performed on all residuals using a Pan and Tompkins QRS detector and the channel with the smoothest FHR time series was selected. Results: The FUSE algorithm performed better than all the individual methods on the training set data. On the validation set, best Challenge scores obtained were E4=29.6. E5=4.67 for events 4-5 respectively using the FUSE method.