Automated Evaluation of Respiratory Signals to Provide Insight Into Respiratory Drive

Abstract

The diaphragm muscle is the primary inspiratory muscle in mammals and is highly active throughout life. The repetitive activation of the diaphragm muscle (and of other muscles driven by central pattern generator activity) presents an opportunity to analyze these physiological data on a per‐event basis. Here we highlight the development and implementation of the Automated Evaluation of Respiratory Signals (AERS) algorithm to detect individual respiratory events across a range of motor behaviors. Because different respiratory‐related signals provide different types of information, the AERS algorithm is designed to be robust regardless of the signal type—e.g., diaphragm electromyography (EMG) and transdiaphragmatic pressure (Pdi). Individual respiratory event data obtained from adult rats were assessed using the AERS algorithm. This analysis revealed diaphragm root‐mean‐square (RMS) EMG and Pdi were lowest during eupnea, increased slightly in response to hypoxia‐hypercapnia (10% O2 ‐ 5% CO2), and increased approximately 3‐fold in response to airway occlusion compared to the eupneic baseline. The coefficient of variation (CV) within subjects was less than 10% across all behaviors for both RMS EMG and Pdi. By contrast, the CV was approximately 30% across subjects. Respiratory rate was increased by ~20 breaths/min during hypoxia‐hypercapnia compared to eupnea, as expected based on previous work from our lab and others. The respiratory rate CV within subjects was less than 10% during eupnea and hypoxia‐hypercapnia; during airway occlusion, the CV was ~17%, likely owing to the small number of respiratory events (2‐5) analyzed for this behavior per animal. For eupnea and hypoxia‐hypercapnia, the CV across subjects was ~20%, whereas during airway occlusion it was ~25%. Overall, our findings validate the mean values across various measures of respiratory motor output compared to standard subject means approach. However, per‐event analyses greatly enhance statistical power, allowing detection of physiologically relevant measures by taking within‐subject variability into account. Additionally, the AERS analyses show that various measures of respiratory variability as well as time‐frequency analyses can provide deeper insights into respiratory drive when accounting for individual respiratory events. These analyses can be effectively applied to a variety of models of respiratory motor unit loss or dysfunction (e.g., spinal cord injury and aging), allowing the extraction of large amounts of data from respiratory signals for more comprehensive assessments. Our findings suggest that the AERS approach may be particularly beneficial in reducing animal numbers in certain types of studies, for the detection of physiological changes reflected in respiratory signals with interventions causing gain or loss of function, and for analyses of respiratory variability, where breath‐to‐breath dynamics are important.