The detection of crackles based on mathematical morphology in spectrogram analysis.

BACKGROUND Crackles are very common abnormal breath sounds in the lung and can be used to diagnose pulmonary diseases. OBJECTIVE In this study, a method is proposed for the detection of adventitious transient sounds from normal breath sounds. METHODS This method automatically recognizes crackles based on the extraction and analysis of spectral information from digitally recorded lung sounds. Various mathematical morphology feature sets were extracted through wavelet spectrogram analysis on pulmonary signals. In order to evaluate the effects of different wavelets types on crackle detection, different wavelets were tested. RESULTS The results showed that the proposed method achieved an 86% accuracy in the detection of crackles. CONCLUSIONS The spectrograms of the crackles in the lung exhibit irregular ellipse image features. For lung sound analysis, this is a useful feature that can be used for the immediate recognition and analysis of crackles.

[1]  Fan Wang,et al.  Automatic detection of microcalcifications using mathematical morphology and a support vector machine. , 2014, Bio-medical materials and engineering.

[2]  Nizamettin Aydin,et al.  Pulmonary crackle detection using time-frequency and time-scale analysis , 2013, Digit. Signal Process..

[3]  L.J. Hadjileontiadis,et al.  Detection of explosive lung and bowel sounds by means of fractal dimension , 2003, IEEE Signal Processing Letters.

[4]  Leontios J. Hadjileontiadis.  A Texture-Based Classification of Crackles and Squawks Using Lacunarity , 2009, IEEE Transactions on Biomedical Engineering.

[5]  Mekki Ksouri,et al.  Wavelets: An efficient tool for lung sounds analysis , 2008, 2008 IEEE/ACS International Conference on Computer Systems and Applications.

[6]  Boualem Boashash,et al.  Time-Frequency Signal Analysis and Processing: A Comprehensive Reference , 2015 .

[7]  Mahesh Pawar,et al.  Analysis of Deformities in Lung Using Short Time Fourier Transform Spectrogram Analysis on Lung Sound , 2011, 2011 International Conference on Computational Intelligence and Communication Networks.

[8]  Wu Xiaoming,et al.  A novel method for feature extraction of crackles in lung sound , 2012, 2012 5th International Conference on BioMedical Engineering and Informatics.

[9]  Shan Hu,et al.  Texture feature extraction based on wavelet transform and gray-level co-occurrence matrices applied to osteosarcoma diagnosis. , 2014, Bio-medical materials and engineering.

[10]  Zhang Hong,et al.  Texture feature extraction based on wavelet transform , 2010, 2010 International Conference on Computer Application and System Modeling (ICCASM 2010).

[11]  Kenneth Sundaraj,et al.  Computer-based Respiratory Sound Analysis: A Systematic Review , 2013 .

[12]  Minghui Du,et al.  HHT based lung sound crackle detection and classification , 2005, 2005 International Symposium on Intelligent Signal Processing and Communication Systems.