Multichannel EHG segmentation for automatically identifying contractions and motion artifacts

In a recent past, several techniques have been developed to analyze the events contained in the electrohysterogram signals (EHG). But, the most of them focused on offline methods. In this study, we use an online method which is developed previously and known by Dynamic Cumulative Sum (DCS). The approach is applied on real EHG signals database through a 4×4 electrodes matrix. For this purpose, two parameters affecting the DCS method, the size of analyzing window and the function detection thresholding value, were first tuned in order to identify the recommended values for ruptures detection from the EHG signals by comparing to the same labeled signals. Furthermore, data fusion of the contraction extracted data has been applied and then analyzed by extracting the contractions detection performance and the performance on electrodes. According to the obtained results, DCS seems to be an encouraging method that could be used for automatic ruptures segmentation. However, further efforts are needed to apply fusion methods to the obtained data from DCS applied on the sixteen electrodes.