Abstract
Lung sounds contribute essential information about a patient’s health. Here, for detecting adventitious sounds of lungs, two sets of eight features each (total 16 features), based on wavelet bi-spectrum (WBS) and power spectrum (WPS), respectively, are proposed. The feature sets are analyzed using five classifiers with one to seven sub-classifier types. A matrix (17X14) of seven evaluation parameters compares the feature sets. Results show that Random Forest, Random Tree, Random committee with WPS and WBS features, LMT with WPS, and Randomizable filter classifier with WBS have shown the best results updating the accuracy obtained in previous researches.
ACKNOWLEDGEMENT
The authors are thankful to Research Centre MCTE MHOW, and IET DAVV University Indore, India, for providing the necessary research infrastructure. Also express gratitude to parent organizations MCTE MHOW, IPS Academy, Institute of Engineering and Science, and SDBCE Indore.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
The author(s) reported there is no funding associated with the work featured in this article.
Nomenclature
a | = | Scaling factor |
b | = | Dilation factor |
Wx (a,b) | = | Continuous wavelet transform |
t | = | time |
ψ(t) | = | Wavelet type complex Morlet |
WBx(a1,a2) | = | Wavelet bi-spectrum |
= | Randomizable Time | |
* | = | Complex conjugate |
ω0 | = | Dimensionless frequency |
LP | = | Local peak |
GP | = | Global peak |
l | = | Number of GPs and LPs |
i | = | Position of the max value of peak |
meanWBxLP | = | Mean of wavelet bi-spectrum over local peaks |
meanPxLP | = | Mean of wavelet power spectrum over local peaks |
stdWBxLP | = | Standard deviation of wavelet bi-spectrum over local peaks |
stdPxLP | = | Standard deviation of waveletpower spectrum over local peaks |
ci | = | Position of the maximum value of the ith GP at the bi-frequency domain |
mAx (ω1, ω2) | = | Average instantaneous wavelet bi-amplitude across the examined time interval Ttotal |
mAbxGPi (ωc1, ωc2) | = | Global peak (GP) value |
ωc1, ωc2 | = | Frequency |
mAbxLPi (ωc1, ωc2) | = | Local peak (LP) value |
DGPi | = | Distance of the Ci from the contour S of the ith GP at the bi-frequency domain. |
Additional information
Notes on contributors
![](/cms/asset/25b533c6-2520-4d8d-8abb-255dee5c72ee/tijr_a_1977188_ilg0001.gif)
Rupesh Dubey
Rupesh Dubey (M'40), FIETE, associate member ISTE, IAENG, IRED, IDES did BE from RGPV University, Bhopal, and received ME from SGSITS Indore both in electronic & telecommunication engineering. Presently pursuing a PhD from Military College of Telecommunication Engineering, Mhow India, DAVV University Indore in the domain of biomedical signal processing. Currently, Head EC IPSA, IES, Indore (An UGC Autonomous Institute), with NBA accredited UG programmes. Research interests are in signal processing, MIMO Systems, HOSA. Published and presented more than 30 research papers in journals and conferences. He has reviewed many conferences papers and journals, articles. He has also registered his patient. He has guided 13 PG scholars.
![](/cms/asset/0eb3cc69-f873-4cdf-a721-dd48418a88f1/tijr_a_1977188_ilg0002.gif)
Rajesh M. Bodade
Rajesh M Bodade (M'55) FIETE, Member IEI, ISTE, BE Elec & Tc Engg from Marathwada SU, Aurangabad, ME Electronics-Sp CT from SRTMU, and PhD E & Tc. SRTM University, Nanded. With 25 years of experience, currently working as professor at FCE, Military College of Telecommunication Engineering, Mhow, Indore, India. Awarded CASMC, GOCCATCMC, for outstanding contribution in teaching and research at MCTE. Research interests include signal processing, pattern recognition, and biometrics. Published and presented 30 research papers in journals and conferences. Authored book Analysis of Iris for Biometric Recognition Systems Springer. He is reviewer for journals like PAMI, PR Letters of Elsevier, IJIST Wiley. Email: [email protected]
![](/cms/asset/c53adfc1-ce52-463a-b405-bf5b15dba4a0/tijr_a_1977188_ilg0003.gif)
Divya Dubey
Divya Dubey (F'40), member IETE, did her MTech degree in instrumentation from DAVV University Indore. During post graduate degree, completed a project from the prestigious IISc, Bangalore. She is currently serving a assistant professor SDBCE Indore. Her research interests are in signal processing, instrumentation, and electronics. Published and presented research papers in journals and conferences. Email: [email protected]