Analysis of central venous pressure (CVP) signals using mathematical methods

Central venous pressure (CVP) is an important clinical parameter for physicians but only the absolute CVP value is typically monitored in the intensive care unit (ICU). In this study, we propose a novel mathematical method to present and analyze CVP signals. A total of 44 suitable samples were chosen from a total of 65 collected in an ICU. Pre-processing of the samples included rate reduction and digital filtering. The statistical features of time and frequency domain, wavelet, and empirical mode decomposition of these signals were extracted. We found no significant difference among the CVP signals regarding sex, smoking, coronary disease, and respiration mode of the samples.