Frequency range extension of spectral analysis of pulse rate variability based on Hilbert-Huang transform

Chia Chi Chang, Tzu-Chien Hsiao, Hung Yi Hsu*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

18 Scopus citations

Abstract

Heart rate variability (HRV) is a well-accepted indicator for neural regulatory mechanisms in cardiovascular circulation. Its spectrum analysis provides the powerful means of observing the modulation between sympathetic and parasympathetic nervous system. The timescale of HRV is limited by discrete beat-to-beat time intervals; therefore, the exploration region of frequency band of HRV spectrum is relatively narrow. It had been proved that pulse rate variability (PRV) is a surrogate measurement of HRV in most of the circumstances. Moreover, arterial pulse wave contains small oscillations resulting from complex regulation of cardiac pumping function and vascular tone at higher frequency range. This study proposed a novel instantaneous PRV (iPRV) measurement based on Hilbert-Huang transform. Fifteen healthy subjects participated in this study and received continuous blood pressure wave recording in supine and passive head-up tilt. The result showed that the very-high-frequency band (0.4-0.9 Hz) varied during head-up tilt and had strong correlation (r = 0.77) with high-frequency band and medium correlation (r = 0.643) with baroreflex sensitivity. The very-high-frequency band of iPRV helps for the exploration of non-stationary autoregulation and provides the non-stationary spectral evaluation of HRV without distortion or information loss.

Original languageEnglish
Pages (from-to)343-351
Number of pages9
JournalMedical and Biological Engineering and Computing
Volume52
Issue number4
DOIs
StatePublished - Apr 2014

Keywords

  • Heart rate variability
  • Hilbert-Huang transform
  • Pulse rate variability

Fingerprint

Dive into the research topics of 'Frequency range extension of spectral analysis of pulse rate variability based on Hilbert-Huang transform'. Together they form a unique fingerprint.

Cite this