Background: The multi-mode modulation is a key feature of sleep EEG. And the short-term fractal property reflects the sympathovagal modulation of heart rate variability (HRV). The properties of EEG and HRV strongly correlated with sleep status and are interesting in clinic diagnosis. New method: 19 healthy female subjects were included for over-night standard polysomnographic study. Hilbert Huang transform (HHT) was used to characterize the temporal features of slow- and fast-wave oscillations decomposed from sleep EEG at different stages. Masking signals were used for solving the mode-mixing problem in HHT. On the other hand, detrended fluctuation analysis (DFA) was used to assess short-term property of HRV denoted as DFA α1, which reflects the temporal activity of autonomic nerve system (ANS). Thus, the dynamic interaction between sleep EEG and HRV can be examined through the relationship between the features of sleep EEG and DFA α1 of HRV. Results: The frequency feature of sleep EEG serves as a good indicator for the depth of sleep during non-rapid eye movement (NREM) sleep, and amplitude feature of fast-wave oscillation is a good index for distinguishing rapid eye movement (REM) from NREM sleep. Comparison with existing method: The relationship between DFA α1 of HRV and the mean amplitude of fast-wave oscillation of sleep EEG affirmed with Pearson correlation coefficient is more significant than the correlation verified by the traditional spectral analysis. Conclusion: The dynamic properties of sleep EEG and HRV derived by EMD and DFA represent important features for cortex and ANS activities during sleep.

Original languageEnglish
Pages (from-to)233-239
Number of pages7
JournalJournal of Neuroscience Methods
Issue number2
Publication statusPublished - Oct 15 2013


  • Detrended fluctuation analysis
  • Heart rate variability
  • Hilbert Huang transform
  • Sleep EEG
  • Sympathovagal modulation

ASJC Scopus subject areas

  • Neuroscience(all)


Dive into the research topics of 'Investigating the interaction between heart rate variability and sleep EEG using nonlinear algorithms'. Together they form a unique fingerprint.

Cite this