Dynamical pattern representation of cardiovascular couplings evoked by head-up tilt test

Danuta Makowiec , Dorota Wejer , Beata Graff , Zbigniew Struzik

Abstract

Shannon entropy (ShE) is a recognised tool for the quantization of the temporal organization of time series. Transfer entropy (TE) provides insight into the dependence between coupled systems. Here, signals are analysed that were produced by the cardiovascular system when a healthy human underwent a provocation test using the head-up tilt (HUT) protocol. The information provided by ShE and TE is evaluated from two aspects: that of the algorithmic stability and that of the recognised physiology of the cardiovascular response to the HUT test. To address both of these aspects, two types of symbolization of three-element subsequent values of a signal are considered: one, well established in heart rate research, referring to the variability in a signal, and a novel one, revealing primarily the dynamical trends. The interpretation of ShE shows a strong dependence on the method that was used in signal pre-processing. In particular, results obtained from normalized signals turn out to be less conclusive than results obtained from non-normalized signals. Systematic investigations based on surrogate data tests are employed to discriminate between genuine properties—in particular inter-system coupling—and random, incidental fluctuations. These properties appear to determine the occurrence of a high percentage of zero values of TE, which strongly limits the reliability of the couplings measured. Nevertheless, supported by statistical corroboration, we identify distinct timings when: (i) evoking cardiac impact on the vascular system, and (ii) evoking vascular impact on the cardiac system, within both the principal sub-systems of the baroreflex loop.
Author Danuta Makowiec (FMPI / ITPA)
Danuta Makowiec,,
- Institute of Theoretical Physics and Astrophysics
, Dorota Wejer (FMPI / IEP)
Dorota Wejer,,
- Institute of Experimental Physics
, Beata Graff
Beata Graff,,
-
, Zbigniew Struzik
Zbigniew Struzik,,
-
Journal seriesEntropy, ISSN 1099-4300, (A 30 pkt)
Issue year2018
Vol20
No4
Pages1-24
Keywords in Englishtime series analysis, dynamic symbolization, pattern representation, information transfer, orthostatic stimulus
DOIDOI:10.3390/e20040235
URL https://www.mdpi.com/1099-4300/20/4/235
Languageen angielski
LicenseJournal (articles only); published final; Uznanie Autorstwa (CC-BY); with publication
Score (nominal)30
ScoreMinisterial score = 30.0, 25-10-2018, ArticleFromJournal
Ministerial score (2013-2016) = 30.0, 25-10-2018, ArticleFromJournal
Publication indicators WoS Impact Factor: 2016 = 1.821 (2) - 2016=1.947 (5)
Citation count*
Cite
Share Share

Get link to the record


* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back