Parameters extracted from arterial pulse waves as markers of atherosclerotic changes: performance and repeatability

<p>Arterial diseases are significant and increasing cause of mortality and morbidity. In this study, we analyze and compare the discrimination capability of different arterial pulse wave (PW) based indices, both earlier proposed and novel ones, for describing the vascular health. The repeatability of the indices is also evaluated. Both volume PWs and dynamic pressure PWs are recorded by using photoplethysmographic and electromechanical film (EMFi) sensors connected to a wireless body sensor network. The study population consists of 82 subjects, 30 atherosclerotic patients and 52 control subjects. In addition, day-to-day variability of the derived indices is studied with 10 test subjects examined on three different days. The results are evaluated in terms of statistical tests and receiver operating characteristic (ROC) curves as well as coefficient of variation (CV) and intra-class correlation coefficient (ICC). Altogether 24 out of the evaluated 40 PW parameters showed statistical differences (p < 0:05 or less) between controls and atherosclerotic patients. Maximum area under curve was 0.88. Most of the indices had ICCs higher than 0.8 and average CVs less than 0.1. The study shows that the amplitude ratios and time intervals between different PW peaks could be a useful additional tool for the detection of atherosclerosis. The results encourage us for further studies in this field. Up to our knowledge, the performance and the repeatability of different PW derived indices have previously not been studied and compared with each other this extensively. Our findings also provide evidence for the utility of PW measurements for the detection of atherosclerotic changes.</p>