Abstract:
Pulmonary crackles are used as indicators for the diagnosis of different pulmonary disorders. Crackles are very common adventitious sounds which have transient characteri...View moreMetadata
Abstract:
Pulmonary crackles are used as indicators for the diagnosis of different pulmonary disorders. Crackles are very common adventitious sounds which have transient characteristics. From the characteristics of crackles such as timing and number of occurrences, the type and the severity of the pulmonary diseases can be obtained. In this study, a method is proposed for crackle detection. In this method, various feature sets are extracted using time-frequency analysis. In order to understand the effect of using different window types in time-frequency analysis in detecting crackles, various types of windows are used such as Gaussian, Blackman, Hanning, Hamming, Bartlett, Triangular and Rectangular. The extracted features both individually and as an ensemble of networks sets are fed into k-Nearest Neighbor classifier. Besides, in order to improve the success of the classifier, prior to the time frequency analysis, frequency bands containing no-crackle information are removed using dual tree complex wavelet transform, which is a shift invariant transform with limited redundancy compared to the conventional discrete wavelet transform. The comparative results of individual feature sets and ensemble of sets, which are extracted using different window types, for pre-processed and non pre-processed data with k-Nearest Neighbor are extensively evaluated and compared.
Date of Conference: 18-20 April 2012
Date Added to IEEE Xplore: 28 May 2012
ISBN Information:
Print ISSN: 2165-0608