New adaptive thresholding-based ECG R-peak detection technique

The QRS complexes (R-peaks) in the electrocardiogram (ECG) signals are essential features as they provide information for the diagnosis of heart diseases. Because detection accuracy is a major concern in clinical practices, most of the detection methods are complex and computationally expensive. However, in ubiquitous healthcare (u-healthcare), the energy consumption is a question; thus, detection time has to be minimized. In this paper, we propose a new intelligent detection algorithm, which minimizes detection time while maintaining a good level of detection performance. Experimental results, in detecting the R-peaks in 30 minutes long records, showed that the proposed algorithm has achieved an Se (sensitivity) of 99.63% and a +P (positive predictivity) of 99.50% outperforming the well-known algorithms from literature.