Onset detection in arterial blood pressure pulses using Empirical Wavelet Transform

In this article, a new algorithm is proposed for the detection of onsets in arterial blood pressure (ABP) signals. The proposed algorithm utilizes Empirical Wavelet Transform (EWT) for pulse onset detection. EWT is a new mathematical tool pursuing a similar goal like Empirical mode decomposition (EMD) is and it explicitly builds an adaptive wavelet filter bank to decompose a given signal into different modes. Arterial blood pressure pulses provide ample pathophysiological information regarding cardiovascular circulation system. Arterial blood pressure pulses analysis is regular practice used to investigate the health status of cardiovascular system. Three different databases (fantasia, Multiparameter Intelligent Monitoring in Intensive Care (MIMIC) and MIT-BIH ploysomnoghaphic database) were considered during the validation and evaluation of the proposed algorithm. The proposed algorithm achieved an average error rate of 0.14%, sensitivity of 99.92% and positive predictivity of 99.92%.