Wheeze type classification using non-dyadic wavelet transform based optimal energy ratio technique

• Published in: Computers in biology and medicine 2019-01, Vol.104, p.175-182
• Main Authors: Ulukaya, Sezer, Serbes, Gorkem, Kahya, Yasemin P.
• Format: Article
• Language: English
• Subjects: Accuracy; Acoustics; Anomalies; Asthma; Chronic obstructive pulmonary disease; Classification; Classifiers; Conflicts of interest; Discrete Wavelet Transform; Discrimination; Humans; Learning algorithms; Lung - physiopathology; Lung diseases; Lung Diseases - physiopathology; Monophonic; Neural networks; Obstructive lung disease; Pathology; Polyphonic; Pulmonary sounds; Representations; Respiratory sounds; Respiratory Sounds - classification; Respiratory Sounds - physiopathology; Signal classification; Signal Processing, Computer-Assisted; State of the art; Support Vector Machine; Support vector machines; Wavelet Analysis; Wavelet transforms; Wheezing
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2018.11.004

Wheezes in pulmonary sounds are anomalies which are often associated with obstructive type of lung diseases. The previous works on wheeze-type classification focused mainly on using fixed time-frequency/scale resolution based on Fourier and wavelet transforms. The main contribution of the proposed method, in which the time-scale resolution can be tuned according to the signal of interest, is to discriminate monophonic and polyphonic wheezes with higher accuracy than previously suggested time and time-frequency/scale based methods. An optimal Rational Dilation Wavelet Transform (RADWT) based peak energy ratio (PER) parameter selection method is proposed to discriminate wheeze types. Previously suggested Quartile Frequency Ratios, Mean Crossing Irregularity, Multiple Signal Classification, Mel-frequency Cepstrum and Dyadic Discrete Wavelet Transform approaches are also applied and the superiority of the proposed method is demonstrated in leave-one-out (LOO) and leave-one-subject-out (LOSO) cross validation schemes with support vector machine (SVM), k nearest neighbor (k-NN) and extreme learning machine (ELM) classifiers. The results show that the proposed RADWT based method outperforms the state-of-the-art time, frequency, time-frequency and time-scale domain approaches for all classifiers in both LOO and LOSO cross validation settings. The highest accuracy values are obtained as 86% and 82.9% in LOO and LOSO respectively when the proposed PER features are fed into SVM. It is concluded that time and frequency domain characteristics of wheezes are not steady and hence, tunable time-scale representations are more successful in discriminating polyphonic and monophonic wheezes when compared with conventional fixed resolution representations. [Display omitted] •A Rational Dilation Wavelet Transform based on peak energy ratio parameter selection method is proposed.•Optimal time-frequency representations of wheeze lung sounds are obtained by tuning Q-factors of the non-dyadic wavelets.•Energy features derived from optimum wavelet decomposition are used in three classification algorithms.•Performance of proposed system is compared with former time-frequency/scale based methods in two cross validation settings.•Superiority of proposed method over Fourier and wavelet based methods is validated in terms of accuracy and run time.