Multichannel Recording and Classifying of Respiratory Sounds

The Auscultation of the lungs with a stethoscope is an important and simple diagnostic method. It provides direct information about the structure (and function) of lung tissue that cannot be gathered by any other simple and noninvasive method. Changes in the quality of lung sounds are often directly correlated to pathological changes in the lungs. Particularly in practical medicine, auscultation is very important and considered a standard method of diagnosing and monitoring pulmonary diseases alongside lung function testing and chest x-rays. The multiplicity of lung sounds that can be observed makes it difficult even for a physician to arrive at a definite diagnosis. Today auscultation sounds are divided into two main groups: respiratory sounds and rales. Both respiratory sounds and rales contain important diagnostic information. Respiratory sounds include vesicular breathing, trachea air-flow sounds, and bronchial breathing. Rales include moist and dry rales, crackling, crepitation, rhonchus, and wheezing. The small selection of lung sounds mentioned above gives an idea of the difficulty of making a clear diagnosis. A recent research project run jointly by Joanneum Research Graz and the Pulmonary Day Care Center at the LKH-GrazWest hospital in Graz, Austria intended to develop a system similar to a medical expert system to help the general practitioner or specialist diagnose lung diseases quickly and clearly. The objective of the first steps of the research project is to lay the foundation for the system by developing new sensors and the recording software as well as to do extensive physicals.