METHODS OF RESPIRATORY DATA ANALYSIS
Abstract (summary)
Methods for evaluating temporal and nontemporal interrelationships among respiratory parameters, which are important to the understanding of breath to breath regulation of respiration, were developed and applied to respiratory data obtained for awake normal adult supine subjects under unstressed conditions using a computerized noninvasive canopy spirometer system. The parameters evaluated were: tidal volume (VT), minute ventilation (VE), mean inspiratory flow (VI), mean expiratory flow (VO), breath duration (TTOT), inspiratory time (TI), total expiratory time (TE), active expiratory time (TA), expiratory dwell (TD), and duty cycle (DC).
The general approach for both the nontemporal and temporal analyses was development of quantitative descriptors of the individual respiratory parameters and examination of the changes in these descriptors among the parameters. Emphasis was placed on investigating groups of respiratory parameters related by defining equations such as; VI=VT/TI and VE=VT/TTOT. A quantitative definition of extremely large or small respiratory parameter values, or outliers, was developed for use in both nontemporal and temporal analyses. A nontemporal interrelationship identified was the concurrence of outliers in VT, TI, and TTOT which allowed large excursions in VT while VI and VE vary within narrower limits.
Short term recurrent patterns, extending over a few breaths, were identified through use of an autocorrelation function histogram technique. Periodicities of 3 to 33 breaths per cycle were identified through comb filtering. Both methods included procedures for handling trends, outliers, and the apparent nonstationarity of respiratory data sequences. The volume related components (VT, VE, VI, VO) exhibited recurrent patterns of lower frequency than the timing components (TI, TE, TA, TD, TTOT, DC). VE showed the strongest recurrent patterns suggesting a sensitive control mechanism for this parameter. Using complex demodulation, a 180 degree phase difference between concurrent periodicities in TI and VI was noted in subjects with low VT-TI correlation coefficients for breaths in the nonoutlier range. . . . (Author's abstract exceeds stipulated maximum length. Discontinued here with permission of author.) UMI
Indexing (details)
Biomedical engineering