Robust arterial blood pressure onset detection method from signal artifacts

Abstract

Arterial blood pressure (ABP) is used in various areas such as brain computer interface and clinical field. The morphological analysis of the ABP signal allows researchers to identify important information such as cardiovascular system and psychopathology. Detection of onset, which is the most important landmark in the ABP waveform, is essential for morphology analysis of ABP. Since the physiological signal is vulnerable to the risk of contamination, the robust onset detection method is needed. This study proposed a pulse onset detection method based on Monte Carlo approach that is robust from artifacts. The 10 cases of ABP signals were analyzed to detect signal onset. When we assessed the time difference from the actual onset, there was an average error of 2.4μs. The results suggested that the proposed method could achieve robustness in pulse detection and facilitated pulse wave analysis using clinical recordings with various artifacts.