A software toolkit for acoustic respiratory analysis
Author(s)Yi, Gina Ann
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
John V. Guttag.
MetadataShow full item record
Millions of Americans suffer from pulmonary diseases. According to recent statistics, approximately 17 million people suffer from asthma, 16.4 million from chronic obstructive pulmonary disease, 12 million from sleep apnea, and 1.3 million from pneumonia - not to mention the prevalence of many other diseases associated with the lungs. Annually, the mortality attributed to pulmonary diseases exceeds 150,000. Clinical signs of most pulmonary diseases include irregular breathing patterns, the presence of abnormal breath sounds such as wheezes and crackles, and the absence of breathing entirely. Throughout the history of medicine, physicians have always listened for such sounds at the chest wall (or over the trachea) during patient examinations to diagnose pulmonary diseases - a procedure also known as auscultation. Recent advancements in computer technology have made it possible to record, store, and digitally process breath sounds for further analysis. Although automated techniques for lung sound analysis have not been widely employed in the medical field, there has been a growing interest among researchers to use technology to understand the subtler characteristics of lung sounds and their potential correlations with physiological conditions. Based on such correlations, algorithms and tools can be developed to serve as diagnostic aids in both the clinical and non-clinical settings.(cont.) We developed a software toolkit, using MATLAB, to objectively characterize lung sounds. The toolkit includes a respiration detector, respiratory rate detector, respiratory phase onset detector, respiratory phase classifier, crackle and wheeze detectors and characterizers, and a time-scale signal expander. This document provides background on lung sounds, describes and evaluates our analysis techniques, and compares our work to approaches in other diagnostic tools.
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2004.Includes bibliographical references (p. 143-147).
DepartmentMassachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology
Electrical Engineering and Computer Science.