Wavelet analysis of voluntary cough sound in patients with respiratory diseases.

Changes in the characteristics of the cough sound may refer to some specific pathological processes and their evolution. In this pilot study we analyzed voluntary cough sound properties in subjects with asthma bronchiale (AB) and chronic obstructive pulmonary disease (COPD) and discriminated them from the control cough sound in healthy subjects. The wavelet transform was used due to a nonstationarity of cough sound recordings. The duration of cough sound was longer during pathological conditions. The longest duration and the highest power of the cough sound were found in COPD. In AB patients, higher frequencies were detected compared with chronic bronchitis and the power of cough sound was shifted to a higher frequency range compared with control coughs. Cough sounds were classified using discriminant analysis with a correct classification rate of about 85-90 %. The method of cough analysis enables an objective quantification of voluntary cough sound with a useful diagnostic and prognostic value.