Analysis of Exercise-Induced Periodic Breathing Using an Autoregressive Model and the Hilbert-Huang Transform

Figures

• 

  Figure 1. A cardiopulmonary exercise testing (CPET) system and breathing patterns. (a) A CPET machine is comprised of a cycle ergometer and pneumotachometer. During the test, the bicycle workload for the patient is increased until maximal exertion is reached. (b–d) Breathing patterns include eupneic, gasping, and periodic breathing (PB). Tidal volume is the volume of air exchange between inhalation and exhalation.

• 

  Figure 2. Measured signals of exercise breath-by-breath ventilation. Very low measurements usually occur when the patient is gasping. In this study, when the ventilation volume was high, these exceptional measurements were observed as noises.

• 

  Figure 3. An example of empirical mode decomposition of most exhausted exercise ventilations (MEE-Ve). The MEE-Ve signals are k ventilations before the peak volume. The number k is 200 in this paper. The illustration is from the analysis of the patient ID: pb0001 which was judged as a periodic breathing (PB) case. All results of the EMD analysis for the PB or non-PB cases can be found at the Github repository (https://github.com/htchu/EpbAnalysis).

• 

  Figure 4. The corresponding instantaneous frequency of the decomposed IMF1-IMF4 (Figure 3).

• 

  Figure 5. Peak Computations of IMFs (Figure 3).

• 

  Figure 6. Statistical significance test for the decomposed IMFs. The IMF5 is below the 95% confidence limit and is therefore considered statistically insignificant.