Power spectral analysis of the diaphragm electromyogram

T. K. Aldrich, J. M. Adams, N. S. Arora, D. F. Rochester

Research output: Contribution to journalArticlepeer-review

26 Scopus citations

Abstract

We studied the power spectrum of the diaphragm electromyogram (EMG) at frequencies between 31 and 246 Hz in four young normal subjects and five patients with chronic obstructive lung disease (COPD). Diaphragm EMGs were analyzed during spontaneous breathing and maximum inspiratory efforts to determine the effect of signal-to-noise ratio on the power spectrum and if treadmill exercise to dyspnea was associated with diaphragm fatigue. We found that the centroid frequencies of the power spectra (fc) were strongly correlated (r = 0.93) with ratios of power at high frequencies to power at low frequencies (H/L) for all subjects. Of the two indices, H/L had the largest standard deviation expressed as a percentage of the mean. The mean values of both of these decreased significantly after exercise, fc from 100.2 to 97.3 and H/L from 1.07 to 0.97. Signal-to-noise ratios were higher in maximal inspiratory efforts and after exercise in normal subjects and higher in COPD patients. The signal-to-noise ratio was correlated negatively with fc and H/L, indicating that these indices of the shape of the power spectrum are influenced by signal strength and noise levels as well as muscle function. We conclude that the fc and H/L index similar qualities of the power spectrum, that they are partially determined by the signal-to-noise ratio, and that, in some cases, exercise to dyspnea is associated with apparently mild diaphragm fatigue.

Original languageEnglish (US)
Pages (from-to)1579-1584
Number of pages6
JournalJournal of Applied Physiology Respiratory Environmental and Exercise Physiology
Volume54
Issue number6
DOIs
StatePublished - 1983
Externally publishedYes

ASJC Scopus subject areas

  • Physiology
  • Endocrinology

Fingerprint

Dive into the research topics of 'Power spectral analysis of the diaphragm electromyogram'. Together they form a unique fingerprint.

Cite this