Computational and experimental models for sound transmission in the pulmonary system and chest to aid in medical diagnosis

Abstract

Acoustic wave propagation in the pulmonary system and torso is simulated by coupling a numerical acoustic boundary element model that predicts sound propagation throughout the solid tissues with a proven comprehensive analytical model for sound wave propagation in the airways of the bronchial tree that is valid up to at least 2 kHz. This approach enables modeling various pathologies that result in structural changes in the lung and/or changes in breath sound source and strength. The model may be also used to predict the resulting acoustic changes measured by acoustic sensors, e.g., stethoscopes, accelerometers, or other skin-contact sensors. Experimental studies in a novel lung phantom model are used to partially validate the computational model. This study focuses on low audible frequencies, i.e., less than 2 kHz. This range encompasses naturally generated respiratory sounds that have been shown to have diagnostic value, as well as externally-introduced vibro-acoustic stimuli used for diagnosis. [Work supported by NIH EB 003286-01.]