Table 2.

Methods developed to pulmonary sounds analysis.

Methodology	Parameters	References
Time-frequency analysis	Gaussien band width, peak frequency, total deflection width, maximal deflection width	[43] (correct classification level: 87,78%)
Time-frequency analysis	Gaussien band width, peak frequency, maximal deflection width	[43] (correct classification level: 90,5%)
Prony modeling	Parameters of the Prony model	[43][57] (correct classification level: 63,89%)
	Autoregressive coefficients	[58][59]
Wavelet transform	Wavelet scale	[43][47] (correct classification level: 93,9%)
	Wavelet transform fractal dimension based	[60]
	Wavelet transform stationary – non stationary	[61]
Fuzzy rule-based system – FST-NST	27 fuzzy rules	[42]
Artificial neural networks	Autoregressive coefficients, wavelet coefficients, crackles’ parameters	[59]
Empirical mode decomposition	Instrinsic mode function : local zero mean oscillating waves obtained by sifting process	[62]