Classification Of Crackle Sound In Lung Using Discrete Wavelet Transform(dwt) And Restricted Boltzmann Machine(rbm)
DOI:
https://doi.org/10.34818/eoe.v7i2.12947Abstrak
Abstract A crackle sound on lung happen because there’s a anomaly on respiratory tract. Crackle sound like a rattling or clicking that happen to be heard when inhaling or exhaling or both phase of breathing. Ascultate is method that use to evaluate abnormalities inside respiratory tract but this method are still a subjective method. There’s a lot of research that using this problem as its goal with using a different kind of method like using features extraction method, one of those is Discrete Wavelet Transform(DWT). Using Wavelet Transform help to separate the crackle feature from lung sound and using a classification to classify it characteristic. the method is easy to use on wave form of data and it used on this final project. The extracted features classify and tested using Restricted Boltzmann Machine(RBM) which resulting 69% as highest accuracy result. Keywords: lung sound, features extraction, RBM, DWT, crackleUnduhan
Referensi
[1] M. A. Tocchetto, A. S. Bazanella, L. Guimaraes, J. L. Fragoso, and A. Parraga, An embedded classifier of lung sounds based on the wavelet packet transform and ANN, vol. 19, no. 3. IFAC, 2014.
[2] S. Ulukaya, G. Serbes, I. Sen, and Y. P. Kahya, “A lung sound classification system based on the rational dilation wavelet transform,” Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS, vol. 2016-Octob, pp. 3745–3748, 2016.
[3] A. Rizal, R. Hidayat, and H. A. Nugroho, “Pulmonary crackle feature extraction using tsallis entropy for automatic lung sound classification,” Proc. 2016 1st Int. Conf. Biomed. Eng. Empower. Biomed. Technol. Better Futur. IBIOMED 2016, pp. 8–11, 2017.
[4] V. I. Quandt, E. R. Pacola, S. F. Pichorim, H. R. Gamba, and M. A. Sovierzoski, “Pulmonary crackle characterization: approaches in the use of discrete wavelet transform regarding border effect, mother-wavelet selection, and subband reduction,” Res. Biomed. Eng., vol. 31, pp. 148–159, 2015.
[5] A. Kandaswamy, C. S. Kumar, R. P. Ramanathan, S. Jayaraman, and N. Malmurugan, “Neural classification of lung sounds using wavelet coefficients,” Comput. Biol. Med., vol. 34, no. 6, pp. 523–537, 2004.
[6] A. Fischer, “Training Restricted Boltzmann Machines,” KI - Künstliche Intelligenz, vol. 29, no. 4, pp. 441–444, 2015.
[7] Y. Li, “CSC321 Tutorial 9: Review of Boltzmann machines and simulated annealing.”
[8] M. Du, F. H. Y. Chan, F. K. Lam, and J. Sun, “Crackle detection and classification based on matched wavelet analysis,” Annu. Int. Conf. IEEE Eng. Med. Biol. - Proc., vol. 4, no. C, pp. 1638–1641, 1997.
[9] S. G. Mallat, “A theory for multiresolution signal decomposition: the wavelet representation,” IEEE Trans. Pattern Anal. Mach. Intell. ( Vol. 11 , Issue 7 , Jul 1989 ), vol. 28, no. 2, pp. 79–85, 2009.
[10] J. M. Nainggolan, “Transformasi Wavelet Diskrit ( Discrete Wavelet Transforms ): Teori dan Penerapan Pada Sistem Daya,” pp. 1–9, 2016.
[11] A. Rizal, R. Hidayat, and H. A. Nugroho, “Comparison of discrete wavelet transform and wavelet packet decomposition for the lung sound classification,” Far East J. Electron. Commun., vol. 17, no. 5, pp. 1065–1078, 2017.
[12] A. Hashemi, H. Arabalibiek, and K. Agin, “Classification of Wheeze Sounds Using Wavelets and Neural Networks,” vol. 11, pp. 127–131, 2011.
[13] C. Altın, “Comparison of Different Time and Frequency Domain Feature Extraction Methods on Elbow Gesture ’ s EMG,” vol. 4138, no. August, pp. 35–44, 2016.
[14] P. Chopra and S. K. Yadav, “Restricted Boltzmann machine and softmax regression for fault detection and classification,” Complex Intell. Syst., vol. 4, no. 1, pp. 67–77, 2018.
[15] “Softmax Regression - mlxtend.” [Online]. Available: http://rasbt.github.io/mlxtend/user_guide/classifier/SoftmaxRegression/#overview. [Accessed: 24-Nov-2019].



