Feature Extraction of Lung Sounds Using Daubechies Wavelet

  Hupendra Devangan,  Nitin Jain,  Ajay Roy

Diagnosis of lung sound using the stethoscope has the many limitation. It depends on the experience and skill of the medical practitioners. Lung sound classification has two parts, first is the feature extraction and second is the classifier used for the classification. Here a biomedical based system is proposed for the feature extraction of lung sound which is done using daubechies wavelet (db10). The features which have been extracted are energy, entropy, standard deviation and variance of the normal lung sound (e.g. vesicular.) and adventitious lung sounds (e.g. crackles and wheezes) and a comparative study has been given between the various features of lung sounds.

Keywords- Auscultation, Adventitious Lung Sound, Wheeze, Crackles, Vesicular, Entropy, Energy, Variance, Standard Deviation, Daubechies Wavelet, Discrete Wavelet Transform.

Unique Identification Number - IJEDR1502228
Page Number(s) - 1394-1400
Pubished in - Volume 3 | Issue 2 | May 2015
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Hupendra Devangan,  Nitin Jain,  Ajay Roy,   "Feature Extraction of Lung Sounds Using Daubechies Wavelet", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.3, Issue 2, pp.1394-1400, May 2015, Available at :http://www.ijedr.org/papers/IJEDR1502228.pdf