Date of Award

Spring 1972

Document Type

Dissertation

Degree Name

Doctor of Engineering Science in Electrical Engineering

Department

Electrical Engineering

First Advisor

Andrew Ulrich Meyer

Second Advisor

Eugene H. Smithberg

Third Advisor

Mauro Zambuto

Fourth Advisor

Raj Pratap Misra

Fifth Advisor

Ruy V. Lourenco

Abstract

The purpose of the investigation is to study the signals involved in the respiratory system and to obtain a model for the dynamics including the activity of the phrenic nerve as the input and the transpulmonary pressure, air flow and the volume as the response.

Experiments were done on anasthetized dogs under natural breathing, with steady state response to various levels of CO2 in the inspired air, and under electrophrenic stimulation. Bipolar electrodes were used at the phrenic nerve, in the cervical region, either for recording the phrenic neurogram or for supramaximal unilateral (or bilateral) stimulation using pulse frequency modulated sine waves or square waves. Both pulse rate and modulating frequency were kept within the physiological range of breathing. The recorded electromyogram of the diaphragm, picked up by surgical insertion of bipolar electrodes, was amplified, rectified and filtered by a modified third order Paynter filter. The ensemble average of the variables over several breaths of the experimental animal was performed. Spectral analysis of the filtered EMG, under both natural breathing and stimulation, was also carried out.

In the present work, force and tension of the diaphragm were not measured. However, on the basis of physiological evidence, a simple first order model, with a time constant of 0.1 second is proposed to represent the relevant EMG-tension dynamics. The diaphragm, approximated as a segment of a hemispherical dome then yields a tension-pressure relationship depending on the radius of curvature of the diaphragm. The resting radius of curvature is represented as a nonlinear function of the resting lung volume and its variation, in turn, depends on the tidal volume and the resting radius of curvature. Finally, a simple first order transfer function is used to represent the pressure-volume dynamics with estimated respiratory resistance and compliance. An analog computer simulation has been performed using the filtered EMG as the input.

The work reveals a correspondence between phrenic neurogram and electromyogram of the diaphragm supported by both spectral analysis and time-domain measurements. On this basis, the electromyogram is considered to represent the neural command. A special technique facilitates the elimination of the contribution of the cardiac spikes to the spectrum of the EMG. The elimination of the cardiac spike was also facilitated by ensemble averaging. The overall model represents the neuromuscular section of the controller in the respiratory feedback system and its response depends on the recorded EMG under various external conditions. Both the input (EMG) and some of the response variables (transpulmonary pressure, volume and flow) can be measured by non-invasive techniques.

Share

COinS