Tampere University of Technology

TUTCRIS Research Portal

Areas under peripheral pulse waves: a potential marker of atherosclerotic changes

Research output: Contribution to journalArticleScientificpeer-review


Original languageEnglish
JournalPhysiological Measurement
Issue number2
Early online date29 Dec 2017
Publication statusPublished - Feb 2018
Publication typeA1 Journal article-refereed


OBJECTIVE: In this study, we propose a method for finding
 atherosclerotic changes based on the ratios of areas under peripheral arterial pulse
 wave (PW) contours and analyze its performance.

APPROACH: The PW signals
 were recorded with force sensors and photoplethysmographic sensors from ankle,
 wrist, cubital fossa, index finger and second toe from 30 atherosclerotic patients
 and 52 control subjects. In addition, day-to-day repeatability of the method
 was studied with 10 test subjects examined on 3 different days. The ratios of
 areas under the PWs were computed and the results were evaluated by means
 of receiver operating characteristic (ROC) analysis, intra-class correlation (ICC)
 coefficient and multiple linear regression analysis.

MAIN RESULTS: Areas under
 ROC curves of 0.802-0.906 were found for different area ratios having statistically
 significant differences between the atheroslerotic group and control groups. ICCs
 over 0.80 were found widely for the beat-by-beat analyzed data and over 0.95 for
 the data based on the averages over different numbers of PWs. Multiple linear
 regression analysis showed linear dependence between the area ratios and age
 and the diagnosis of atherosclerosis.

SIGNIFICANCE: Our findings may facilitate
 development of novel diagnostic approaches and preventive strategies against
 cardiovascular disorders. However, further studies are needed to confirm the
 results. The presented study demonstrates the potential of arterial PW analysis
 in finding vascular abnormalities.&#13.


  • Journal Article

Publication forum classification

Field of science, Statistics Finland

Downloads statistics

No data available