(3.237.20.246) 您好!臺灣時間:2021/04/15 11:01
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:黃詠揚
研究生(外文):Yung-Yang Huang
論文名稱:在吵雜環境下完善心肺音分離之探討
論文名稱(外文):A Study on Post-filtering Technique for Robust Heart Sound and Lung Sound Separation in Noisy Environments
指導教授:呂學士
指導教授(外文):Shey-Shi Lu
口試委員:游世安孟慶宗彭盛裕孫台平
口試委員(外文):Shi-An YuChin-Chun MengSheng-Yu PengTai-Ping Sun
口試日期:2016-07-25
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:電子工程學研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2016
畢業學年度:104
語文別:英文
論文頁數:76
中文關鍵詞:心音肺音心肺音分離演算法生理訊號的除噪演算法
外文關鍵詞:Heart SoundLung SoundHeart Lung Sound SeparationSeparation AlgorithmPhysiological Denoise Algorithm
相關次數:
  • 被引用被引用:0
  • 點閱點閱:227
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
中文摘要

隨著全世界人口的高齡化,遠端照護系統也越來越重要,遠端照護使老化人口可以較為方便的接受醫療,而生理訊號的監控亦是遠端照護相當重要的一環,本篇論文主要探討的生理訊號為心音及肺音。
傳統上來講,病患去醫院就診,醫生會使用聽診器來聆聽心音以及肺音,並以聆聽的方式觀察有無異樣,主要依據醫生的經驗進行較為主觀的分析。而在遠端照護系統當中,醫療人員可能會需要取得使用者的心音及肺音以進行分析,病患在家中需要自行選擇聽診器放置的位置,以獲得心音及肺音,但是往往病患沒有這方面經驗,不知道該如何擺放聽診器於正確的位置,我們的情境是在家中的病患可以任意擺放多個聽診器在胸腔上,因為心音及肺音的位置極為相近,如果沒有放置到正確的位置,往往會錄到心肺音的混合訊號,此外,使用者在家中有可能受到環境雜訊的影響,這篇論文要解決的問題有兩個,分離混合的心肺音訊號以得到乾淨的心音以及肺音,以及除去背景噪音,透過我們分離以及除噪的演算法,可以得到乾淨的心音以及肺音,再把這兩種生理訊號回傳回醫院以進行分析。
本篇論文主要在研究各種心肺音分離以及除噪的演算法,並分析各種演算法的優缺點。


ABSTRACT

Ageing population is a commonly observed phenomenon in most developed countries. Telemedicine system is becoming increasingly important due to the aging population around the world. In a telemedicine system, the monitoring of physiological signal is an important part. This thesis emphasizes on the heart sound and lung sound.
Traditionally, patient will go to a hospital to examine their heart function or lung function. Doctor in the hospital use a stethoscope to listen to heart sound or lung sound and analyze the patients’ heart function or lung function mainly base on subjective consciousness. In a telemedicine system, when medical personnel need heart sound or lung sound, patients in their own houses need to place the stethoscope. However, patients might lack of the experiences of placing the stethoscopes in the right positions. In our scenario, patients in their houses will have multiple stethoscopes and all the thing patients need to do is place these stethoscopes randomly on the chest. However, because of the position of heart and lung which is very close from each other, we might record the mixture of heart sound and lung sound. Besides, we still have another problem that patients’ house might be in a noisy environment. Therefore, this thesis mainly deal with these two problems. After our algorithm to separate the mixture of heart sound and lung sound into heart sound and lung sound. After using separation algorithm, we use the denoise algorithm to deal with the problem of noisy environments. After the denoise algorithm, we can get the pure heart sound and lung sound and send both of these signals back to hospitals to analyze.
The thesis put emphasis on the separation algorithms and denoise algorithms and also analyze the advantages and the disadvantages of these algorithms.


CONTENTS
中文摘要 I
ABSTRACT II
CONTENTS IV
LIST OF FIGURES VIII
LIST OF TABLE XII
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Thesis Organization 3
Chapter 2 Heart Sound and Lung Sound 5
2.1 Introduction of Heart Sound and Lung Sound 5
2.1.1 Heart Sound 5
2.1.2 Lung Sound 6
2.2 Measurement Difficulties 6
Chapter 3 Algorithm and Analysis 8
3.1 Independent Component Analysis 8
3.1.1 Introduction of Independent Component Analysis 8
3.1.2 Definition of Independent Component Analysis 12
3.1.3 Definition of Independence and the restrictions using ICA 13
3.1.4 Introduction of measures of non-Gaussianity 16
3.1.5 Kurtosis 16
3.1.6 Negentropy 17
3.1.7 Introduction of Preprocessing for ICA 20
3.1.8 Centering 20
3.1.9 Whitening 21
3.1.10 Ambiguities of ICA 22
3.1.11 FastICA 23
3.1.12 FastICA for one unit 23
3.1.13 FastICA for several unit 24
3.1.14 Flow path of the FastICA algorithm 25
3.2 Non-negative Matrix factorization (NMF) 26
3.2.1 Introduction of Non-negative Matrix factorization 26
3.2.2 Cost function 27
3.2.3 Multiplicative update rules 28
3.3 Deep Denoising Autoencoder (DDAE) 29
3.3.1 Introduction of autoenconder 29
3.3.2 Deep denoise autoencoder 31
3.4 Generalized maximum a posteriori spectral amplitude estimation (GMAPA) 32
3.4.1 Introduction 32
3.4.2 Implementation 33
Chapter 4 Experiment Methods and Results 37
4.1 Experiments Using Independent Component Analysis 37
4.1.1 Separation Experiments 37
4.1.2 Separation Experiments with White Gaussian Noise(WGN) 45
4.2 Experiments Using Non-negative matrix factorization 52
4.2.1 Separation Experiments 52
4.2.2 Separation Experiments with White Gaussian Noise(WGN) 58
4.3 Experiments Using Deep Denoise Autoencoder(DDAE) 65
4.3.1 Separation Experiments 65
Chapter 5 Conclusion and Future Work 69
5.1 Conclusion 69
5.2 Future Work 70
Chapter 6 Reference 73



Reference

[1]M. A. Hein,” Telemedicine: an important force in the transformation of healthcare, “Report of US Department of Commerce, International Trade Administration, 2009.
[2]https://en.wikipedia.org/wiki/Telemedicine
[3]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,” in Proc. 2006 Annual International Conference of the IEEE EMBS(EMBC), pp. 5708-5711, 2006.
[4]https://en.wikipedia.org/wiki/Heart_sounds
[5]M. Sarkar, I. Madabhavi, N. Niranjan, and M. Dogra,”Auscultation of the respiratory system,” Annals of thoracic medicine, 2015.
[6]A. Mondal, P. S. Bhattacharya, and G. Saha, “Reduction of heart sound interference from lung sound signals using empirical mode decomposition technique,” Journal of Medical Engineering and Technology, vol. 35, no.6-7, pp. 344-353, Oct. 2011.
[7]T. Tsalaile and S. Sanei, “Separation of heart sound signal from lung sound signal by adaptive line enhancement,” in Proc. 15th European Signal Processing Conf. (EUSIPCO), pp. 1231–1235, 2007.
[8]L. J. Hadjileontiadis and S. M. Panas, “Adaptive reduction of heart sounds from lung sounds using fourth-order statistics,” IEEE Transactions on Biomedical Engineering, vol. 44, no. 7, pp. 642–648, 1997.
[9]M. Kompis and E. Russi, “Adaptive heart-noise reduction of lung sounds recorded by a single microphone,” in Proc. 1992 Annual International Conference of the IEEE EMBS(EMBC), pp. 691-692, 1992.
[10]L. J. Hadjileontiadis and S. M. Panas, “Separation of discontinuous adventitious sounds from vesicular sounds using a wavelet-based filter,” IEEE Transactions on Biomedical Engineering, vol. 44, no. 12, pp. 1269 –1281, Dec. 1997.
[11]I. Hossain and Z. Moussavi, “An overview of heart-noise reduction of lung sound using wavelet transform based filter,” in Proc. 2003 Annual International Conference of the IEEE EMBS(EMBC), pp. 458-461, 2003.
[12]T. H. Falk and W. Y. Chan, “Modulation filtering for heart and lung sound separation from breath sound recordings,” in Proc. 2008 Annual International Conference of the IEEE EMBS(EMBC), pp. 1859-1862, 2008.
[13]C.-S. Lin and E. Hasting, “Blind sources separation of heart and lung sounds based on nonnegative matrix factorization,” in Proc. ISPACS, pp. 731-736, 2013.
[14]Vincent, P., Larochelle, H., Lajoie, I., Bengio, Y., Manzagol, P.,“Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion,” Journal of Machine Learning Research, 11(Dec): 3371-3408, 2010.
[15]A. Hyvarinen, J. Karhunen, and E. Oja, Independent Component Analysis, John Wiley & Sons, 2001.
[16]A. Hyvärinen and E. Oja, Independent Component Analysis: Algorithms and Applications, Neural Networks, 2000.
[17]K. H. Knuth,”Bayesian source separation and localization,” in Proc. SPIE''s International Symposium on Optical Science, Engineering, and Instrumentation, pp. 147-158, 1998.
[18]A. Mondal, P. S. Bhattacharya, and G. Saha, “Null Space Component Analysis for Noisy Blind Source Separation,” Signal Processing, vol. 109, pp. 301-316, April 2015.
[19]Y. Tsao and Y.-H. Lai, “Generalized maximum a posteriori spectral amplitude estimation for speech enhancement,” Speech Communication, 2015.
[20]https://en.wikipedia.org/wiki/Heart_sounds
[21]https://en.wikipedia.org/wiki/Respiratory_sounds
[22]Smaragdis, Paris, et al. "Static and dynamic source separation using nonnegative factorizations: A unified view." IEEE Signal Processing Magazine31.3 (2014): 66-75.
[23]https://en.wikipedia.org/wiki/Autoencoder
[24]Lu, Xugang, et al. "Speech enhancement based on deep denoising autoencoder." Interspeech. 2013.
[25]Y. Ephraim and D. Malah, “Speech enhancement using a minimum mean-square error short-time spectral amplitude estimator,” IEEE Transactions, Acoustics Speech and Signal Processing, vol. ASSP-32, pp. 1109–1121, 1984.
[26]J. Chen, Fundamentals of Noise Reduction in Springer Handbook of Speech Processing, Chapter 43, Springer, 2008.
[27]Littmann Company. Heart and Lung Sounds: 3M Littmann Stethoscopes [Online]. Available:http://solutions.3m.com/wps/portal/3M/enUS/3M-Littmann/stethoscope/littmann-learning-institute/heart-lung-sounds/
[28]M. T. Pourazad, Z. Moussavi, F. Farahmand and R. K. Ward, “Heart sounds separation from lung sounds using independent component analysis,” in Proc. 2006 Annual International Conference of the IEEE EMBS(EMBC), pp. 2736-2739, 2006.


QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔