Semester

Summer

Date of Graduation

2004

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Mark Jerabek

Committee Co-Chair

Dave Frazer.

Abstract

Human studies indicate that cough sound and flow analysis may be useful for diagnosing pulmonary abnormalities. The purpose of this study was to evaluate an animal model for cough sound and flow analysis. A system was designed to expose guinea pigs to aerosols of citric acid (0.39M) and record resulting coughs at different stages of chemically induced specific airway resistance (sRAW). Coughs were divided into three categories (low sRAW, n = 113; moderate sRAW, n = 143; high sR AW, n = 93). 124 cough sound parameters were derived from the analysis of the sound pressure waves recorded during the cough. A principal component analysis was performed on the acquired data, and the resulting parameters were used to train a single neuron feed-forward back propagation neural network. The classification system was able to correctly discriminate between members of the high and low airway constriction groups with an accuracy of 0.946 and a sensitivity and specificity of 0.893.

Share

COinS