Self-similarity in physiological time series

New perspectives from the temporal spectrum of scale exponents

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Most physiological time series have self-similar properties which reflect the functioning of physiological control mechanisms. Self-similarity is usually assessed by detrended fluctuation analysis (DFA) assuming that mono- or bi-fractal models generate the self-similar dynamics. Our group recently proposed a new DFA approach describing self-similarity as a continuous temporal spectrum of coefficients, thus not assuming that "lumped-parameter" fractal models generate the data. This paper reviews the rationale for calculating a spectrum of DFA coefficients and applies this method on datasets of signals whose self-similarity has been extensively studied in the past. The first dataset consists of six electroencephalographic (EEG) derivations collected in a healthy volunteer. The second dataset consists of cardiac intervals and diastolic blood pressures recorded in 60 volunteers at different levels of cardiac sympatho/vagal balance. Results reveal the limits of the traditional "lumped-parameter" approach, and provide information on the role of autonomic outflows in determining cardiovascular self-similarity.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages164-175
Number of pages12
Volume7548 LNBI
DOIs
Publication statusPublished - 2012
Event8th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2011 - Gargnano del Garda, Italy
Duration: Jun 30 2011Jul 2 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7548 LNBI
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other8th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2011
CountryItaly
CityGargnano del Garda
Period6/30/117/2/11

Fingerprint

Self-similarity
Fractals
Time series
Exponent
Fractal
Fluctuations
Cardiac
Blood pressure
Blood Pressure
Coefficient
Interval
Model

Keywords

  • Detrended fluctuation analysis
  • EEG
  • Fractals
  • Heart rate variability

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Castiglioni, P. (2012). Self-similarity in physiological time series: New perspectives from the temporal spectrum of scale exponents. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7548 LNBI, pp. 164-175). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7548 LNBI). https://doi.org/10.1007/978-3-642-35686-5_14

Self-similarity in physiological time series : New perspectives from the temporal spectrum of scale exponents. / Castiglioni, Paolo.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7548 LNBI 2012. p. 164-175 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7548 LNBI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Castiglioni, P 2012, Self-similarity in physiological time series: New perspectives from the temporal spectrum of scale exponents. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 7548 LNBI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7548 LNBI, pp. 164-175, 8th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2011, Gargnano del Garda, Italy, 6/30/11. https://doi.org/10.1007/978-3-642-35686-5_14
Castiglioni P. Self-similarity in physiological time series: New perspectives from the temporal spectrum of scale exponents. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7548 LNBI. 2012. p. 164-175. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-35686-5_14
Castiglioni, Paolo. / Self-similarity in physiological time series : New perspectives from the temporal spectrum of scale exponents. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7548 LNBI 2012. pp. 164-175 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{5bf4bfa7433347e5a7e261adb845ae01,
title = "Self-similarity in physiological time series: New perspectives from the temporal spectrum of scale exponents",
abstract = "Most physiological time series have self-similar properties which reflect the functioning of physiological control mechanisms. Self-similarity is usually assessed by detrended fluctuation analysis (DFA) assuming that mono- or bi-fractal models generate the self-similar dynamics. Our group recently proposed a new DFA approach describing self-similarity as a continuous temporal spectrum of coefficients, thus not assuming that {"}lumped-parameter{"} fractal models generate the data. This paper reviews the rationale for calculating a spectrum of DFA coefficients and applies this method on datasets of signals whose self-similarity has been extensively studied in the past. The first dataset consists of six electroencephalographic (EEG) derivations collected in a healthy volunteer. The second dataset consists of cardiac intervals and diastolic blood pressures recorded in 60 volunteers at different levels of cardiac sympatho/vagal balance. Results reveal the limits of the traditional {"}lumped-parameter{"} approach, and provide information on the role of autonomic outflows in determining cardiovascular self-similarity.",
keywords = "Detrended fluctuation analysis, EEG, Fractals, Heart rate variability",
author = "Paolo Castiglioni",
year = "2012",
doi = "10.1007/978-3-642-35686-5_14",
language = "English",
isbn = "9783642356858",
volume = "7548 LNBI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "164--175",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Self-similarity in physiological time series

T2 - New perspectives from the temporal spectrum of scale exponents

AU - Castiglioni, Paolo

PY - 2012

Y1 - 2012

N2 - Most physiological time series have self-similar properties which reflect the functioning of physiological control mechanisms. Self-similarity is usually assessed by detrended fluctuation analysis (DFA) assuming that mono- or bi-fractal models generate the self-similar dynamics. Our group recently proposed a new DFA approach describing self-similarity as a continuous temporal spectrum of coefficients, thus not assuming that "lumped-parameter" fractal models generate the data. This paper reviews the rationale for calculating a spectrum of DFA coefficients and applies this method on datasets of signals whose self-similarity has been extensively studied in the past. The first dataset consists of six electroencephalographic (EEG) derivations collected in a healthy volunteer. The second dataset consists of cardiac intervals and diastolic blood pressures recorded in 60 volunteers at different levels of cardiac sympatho/vagal balance. Results reveal the limits of the traditional "lumped-parameter" approach, and provide information on the role of autonomic outflows in determining cardiovascular self-similarity.

AB - Most physiological time series have self-similar properties which reflect the functioning of physiological control mechanisms. Self-similarity is usually assessed by detrended fluctuation analysis (DFA) assuming that mono- or bi-fractal models generate the self-similar dynamics. Our group recently proposed a new DFA approach describing self-similarity as a continuous temporal spectrum of coefficients, thus not assuming that "lumped-parameter" fractal models generate the data. This paper reviews the rationale for calculating a spectrum of DFA coefficients and applies this method on datasets of signals whose self-similarity has been extensively studied in the past. The first dataset consists of six electroencephalographic (EEG) derivations collected in a healthy volunteer. The second dataset consists of cardiac intervals and diastolic blood pressures recorded in 60 volunteers at different levels of cardiac sympatho/vagal balance. Results reveal the limits of the traditional "lumped-parameter" approach, and provide information on the role of autonomic outflows in determining cardiovascular self-similarity.

KW - Detrended fluctuation analysis

KW - EEG

KW - Fractals

KW - Heart rate variability

UR - http://www.scopus.com/inward/record.url?scp=84871553910&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871553910&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-35686-5_14

DO - 10.1007/978-3-642-35686-5_14

M3 - Conference contribution

SN - 9783642356858

VL - 7548 LNBI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 164

EP - 175

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -