The Effect of High-Frequency Artifacts in Arterial Blood Pressure Waveforms on Pressure Reactivity Index

Artifact detection algorithms are gaining increasing attention in physiological signal analysis. One of the secondary parameters prone to be affected by artifacts is the Pressure reactivity index (PRx), the moving Pearson’s correlation coefficient between arterial blood pressure (ABP) and intracranial pressure used to assess cerebrovascular reactivity in traumatic brain injury patients. The aim of this study is to analyze the effect of high-frequency noise in the ABP waveform on PRx calculation. Based on the Short-Time Fourier Transform of real ABP signal with noise, noise frequency range 5–25 Hz was selected, modeled, and inserted into undisturbed physiological ABP waveform. Subsequently calculated PRx from physiological undisturbed and noisy ABP signal demonstrated the averaged median absolute deviation changed from 0.1129 to 0.1123, respectively. The study’s results suggest that although the high-frequency artifacts comply with the standard definition of an artifact, their effect on PRx value is minimal, and detection or elimination of the noise can be likely omitted in the algorithms.