Expanded quantitative models for assessment of respiratory diseases and monitoring

The industrial and demographic expansion and associated increased exposure to pollutants continue to be critical factors contributing to the development of respiratory and cardiovascular diseases. Specifically, this paper builds on previously developed quantitative models for assessment of respiratory disorders utilizing acoustical characterization, as auscultation is a primary method used in initial assessment of respiratory and cardiovascular functions. Applicable techniques used in the speech processing domain were utilized to evaluate lung sound signals obtained with a digital stethoscope. Utilization of more sensitive electronic stethoscopes and application of quantitative signal analysis methods offer opportunities for improved diagnosis in children and overall patient monitoring. Reported methodology is based on expanded Gaussian Mixed Models (GMM). These expanded models provide significantly increased levels of peculiar respiratory signal identification reaching over the 92% level, although it is accomplished at higher computational demand. This approach allows broader quantitative analysis, identification and monitoring of respiratory disorders in general.