Adaptive Threshold for QRS Complex Detection Based on Wavelet Transform.

This paper presents a novel approach, adaptive threshold algorithm, for QRS complex detection. It is based on two-to-four scales in COIFLET wavelet transform. Two initial thresholds are defined as up-limited-threshold (ULT) and down-limited-threshold (DLT) at the beginning of the detection. The system then automatically adjusts the values of the threshold to achieve optimal QRS detection. Simulation work shows that adaptive threshold algorithm can give more accurate results than fixed algorithms. By means of weights of missing and fault detection, ULT and DLT are adapted to get more accurate detecting results. More important for this algorithm is that it can increase the recognition efficiency greatly.