研究生: |
鄭立豐 Erwin - Hasting |
---|---|
論文名稱: |
基於非負矩陣分解法之肺心音盲源分離 Blind Source Separation of Heart and Lung Sounds Based on Nonnegative Matrix Factorization |
指導教授: |
林敬舜
Ching-Shun Lin |
口試委員: |
吳乾彌
Chen-Mie Wu 林淵翔 Yuan-Hsiang Lin 陳維美 Wei-Mei Chen 王煥宗 Huan-Chun Wang |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 電子工程系 Department of Electronic and Computer Engineering |
論文出版年: | 2013 |
畢業學年度: | 101 |
語文別: | 英文 |
論文頁數: | 66 |
中文關鍵詞: | Nonnegative matrix factorization 、Heart and lungs sounds 、Blind source separation |
外文關鍵詞: | Nonnegative matrix factorization, Heart and lungs sounds, Blind source separation |
相關次數: | 點閱:120 下載:5 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
Lung sound (LS) brings valuable information for lung status and respiratory analysis. However, the interference of heart sound (HS) usually occurs and raises confusion on pathological state during the LS recording. To solve this question, separation of HS and LS from mixed heart-lung sounds (HLS) has become one of major issues in the biomedical research. A novel approach based on nonnegative matrix factorization (NMF) as one of blind source separation (BSS) techniques is proposed. In this paper, the chosen mixed HLS signal is brought to the time-frequency domain and forms a multivariate data stationary time series. This multivariate data are then processed as another data representation by constant $Q$ transform, which is well known as log-frequency short-time Fourier transform (STFT). The result of log-frequency STFT is then used as the input pattern of NMF. The average performance based on heart noise or interference reduction percentage (HNRP) for quantitative evaluation of the proposed NMF-based approach is above 80% for the normal LS signal and 90% for the abnormal LS which also better than the directly applied NMF. Another advantage provided by NMF is it only requires single channel as input signal instead of multichannel which is usually required by other BSS methods.
Lung sound (LS) brings valuable information for lung status and respiratory analysis. However, the interference of heart sound (HS) usually occurs and raises confusion on pathological state during the LS recording. To solve this question, separation of HS and LS from mixed heart-lung sounds (HLS) has become one of major issues in the biomedical research. A novel approach based on nonnegative matrix factorization (NMF) as one of blind source separation (BSS) techniques is proposed. In this paper, the chosen mixed HLS signal is brought to the time-frequency domain and forms a multivariate data stationary time series. This multivariate data are then processed as another data representation by constant $Q$ transform, which is well known as log-frequency short-time Fourier transform (STFT). The result of log-frequency STFT is then used as the input pattern of NMF. The average performance based on heart noise or interference reduction percentage (HNRP) for quantitative evaluation of the proposed NMF-based approach is above 80% for the normal LS signal and 90% for the abnormal LS which also better than the directly applied NMF. Another advantage provided by NMF is it only requires single channel as input signal instead of multichannel which is usually required by other BSS methods.
[1] H. Pasterkamp, S. S. Kraman, and G. R. Wodicka, "Respiratory sounds: Advances beyond the stethoscope," Amer. J. Respir. Crit. Care Med., vol. 156, pp. 974-987, 1997.
[2] N. Gavriely, M. Nissan, A. H. Rubin, and D. W. Cugell, "Spectral characteristics of chest wall breath sounds in normal subjects" Thorax, vol. 50, no. 12, pp. 12921300, Dec. 1995.
[3] V. Gross, A. Dittmar, T. Penzel, F. Schuttler, and P. V. Wichert, "The relationship between normal lung sounds, age, and gender," Amer. J. Respir. Crit. Care Med., vol. 162, no. 3, pp. 905-909, 2000.
[4] A.A. Luisada, "The areas of auscultation and the two main heart sounds," Med. Times, vol. 92, pp. 811, Jan. 1964.
[5] P. J. Arnott, G. W. Pfeiffer, and M. E. Tavel, "Spectral analysis of heart sounds: Relationships between some physical characteristics and frequency spectra of first and second heart sounds in normals and hypertensives," J Biomed. Eng., vol. 6, no. 2, pp. 121-128, Apr. 1984.
[6] A. Mondal, P. S. Bhattacharya, and G. Saha, "Reduction of heart sound interference from lung sound signals using empirical mode decomposition technique," J. Med. Eng. Tech., vol. 35, no. 6-7, pp. 344-353, Oct. 2011.
[7] L. J. Hadjileontiadis and S. M. Panas, "Adaptive reduction of heart sounds from lung sounds using fourth-order statistics," IEEE Trans. Biomed. Eng., vol. 44, no. 7, pp. 642-648, 1997.
[8] T. Tsalaile and S. Sanei, "Separation of heart sound signal from lung sound signal by adaptive line enhancement," Eur. Sig. Pro. Conf. pp. 1231-1235, 2007.
[9] T. H. Falk and W. Y. Chan, "Modulation filtering for heart and lung sound separation from breath sound recordings," IEEE Intl. Conf. Eng. Med. Biol. Soc., pp. 1859-1862, Aug. 2008.
[10] J. C. Chien, M. C. Huang, Y. D. Lin, and F. C. Chong, "A study of heart sound and lung sound separation by independent component analysis technique," Proc. IEEE Intl. Conf. Eng. Med. Biol. Soc., pp. 5708-5711, Aug. 30 - Sep. 3, 2006.
[11] F. Ayari, M. Ksouri, and A. T. Alouani, "Lung sound extraction from mixed lung and heart sounds FASTICA algorithm," IEEE Mediteran Electrotechnical Conf., pp. 339-342, Mar. 25-28, 2012.
[12] B. Makkiabadi, D. Jarchi, and S. Sanei, "A new time domain convolution BSS of heart and lung sounds," IEEE Intl. Conf. Acoustics, Speech, and Signal Processing, pp. 605-608, Mar. 25-30, 2012.
[13] A. Cichocki et al., Nonnegatve Matrix and Tensor Factorizations, United Kingdom: John Wiley and Sons Ltd, 2009.
[14] O. Okun and H. Priisalu. "Fast nonnegative matrix factorization and its application for protein fold recognition." EURASIP Journal on Applied Signal Processing, p. 62, 2006.
[15] D. D. Lee and H. S. Seung, "Algorithms for non-negative matrix factorization," Adv. Neu. Info. Pro. Sys., Cambridge, MA, MIT Press, pp. 556-562, 2001.
[16] T. D. Hien, D. V. Tuan, and P. V. At, "Additive update algorithm for nonnegative matrix factorization," Intl. J. Comp. Sci. Info. Sec., vol. 10, no. 9, Sep. 2012.
[17] J. Eggert and E. Korner, "Sparse coding and NMF," Proc. IEEE Intl. Joint Conf. Neural Networks, vol. 2, pp. 2529-2533, 2004.
[18] W. Liu, K. Yuan, T. Wang, and S. Chen. "An improved consensus clustering for nonnegative matrix factorization in molecular cancer class discovery," In Chinese Conf. Pattern Recognition, pp. 1-4, 2010.
[19] J. C. Brown and M. S. Puckette, "An efficient algorithm for the calculation of a constant Q transform," Amer. J. Acous. Soc., vol. 92, no. 5, pp. 2698-2701, 1992.
[20] J. C. Brown, "Calculation of a constant Q spectral transform," Amer. J. Acous. Soc., vol. 89, no. 1, pp. 425-434, 1991.
[21] D. Ellis. (2004). "Spectrograms: Constant-Q(log-frequency) and conventional(linear)" [Online]. Available: http://www.ee.columbia.edu/ln/labrosa/matlab/sgram/.
[22] Littmann Company. "Heart and Lung Sounds: 3M Littmann Stethoscopes" [Online]. Available: http://solutions.3m.com/wps/portal/3M/en_US/3M-Littmann/stethoscope/littmann-learning-institute/heart-lung-sounds/.
[23] University of Michigan Heart Sound and Murmur Library. "University of Michigan Heart Sound and Murmur Library" [Online]. Available: http://www.med.umich.edu/lrc/psb/heartsounds/.
[24] H. Pasterkamp. (2008). "R.A.L.E. repository" [Online]. Available: http://www.RALE.ca.
[25] P. Bentley, G. Nordehn, M. Coimbra, and S. Mannor. "The PASCAL Classifying Heart Sounds Challenge 2011 (CHSC2011) Results" [Online]. Available: http://www.peterjbentley.com/heartchallenge/index.html.
[26] UNMC. \Breath sounds" [Online]. Available: http://darwin.unmc.edu/hnp/m2/breathsounds.htm.
[27] D. Arnall. (2002). "Pulmonary Breath Sounds" [Online]. Available: http://jan.ucc.nau.edu/~daa/heartlung/breathsounds/contents.html.
[28] S. Reichert, R. Gass, C. Brandt, and E. Andres, "Analysis of respiratory sounds: state of the art," Clinical Medicine: Circulatory, Respiratory and Pulmonary Medicine, vol. 2, pp. 45-58, 2008.
[29] I. Hossain and Z. Moussavi, "An overview of heart noise reduction of lung sound using wavelet transform based filter," IEEE Intl. Conf. Eng. Med. Biol. Soc., pp. 458-461, Sep. 17-21, 2003.