Cluster analysis of respiratory time series

Area of Science:

• Physiology
• Respiratory Control

Background:

• The respiratory control system involves complex interactions of multiple variables.
• It is hypothesized that the controller simplifies by manipulating only a few key variables.

Purpose of the Study:

• To investigate the underlying simplicity of respiratory control.
• To determine which respiratory variables are primarily manipulated by the controller.

Main Methods:

• Anesthetized dogs were subjected to CO2 breathing and carotid sinus perfusion.
• Time series data for minute ventilation (VI), tidal volume (VT), breathing period (TT), inspiratory duration (TI), expiratory duration (TE), and PACO2 were collected breath-by-breath.
• Cluster Analysis and Factor Analysis were used to analyze the data.

Main Results:

• Respiratory time series characteristics clustered into magnitude, speed, variability, and relative change.
• VT and TI showed similar time courses for magnitude and relative change.
• TT and TE consistently clustered together across all characteristics.
• Factor analysis revealed latent factors associated with [TT-TE] and PACO2.

Conclusions:

• The respiratory control system appears to operate with a reduced set of independent variables.
• Breathing period (TT) and expiratory duration (TE) are key, consistently regulated variables.
• Specific latent factors, particularly related to [TT-TE], significantly explain respiratory control dynamics.