臺灣博碩士論文加值系統

傳統心音聽診是醫生使用聽診器做初步診斷，進一步利用心臟超音波(Echo-cardiography)評估病人的心臟結構，其缺點為後者須以高成本購置心臟超音波，加上每次量測病人的時間太長，造成耗費許多無形的金錢與時間。本論文與高雄榮總及高雄長庚醫院醫生合作，使用本實驗室現有心音計裝置，採集一百零八筆有效樣本的臨床心音訊號，進行訊號分析。
透過高單價心臟超音波報告加以比對，將臨床心音數據分成正常與異常兩大類，經由快速傅立葉轉換(FFT)演算法，把量測到的心音與心電訊號從時域轉換到頻域，利用帶通濾波器(BPF)從特定頻帶中找出兩者的差異性。並且利用心音圖(PCG)裝置與心電圖(ECG)的生理資訊比對功能，觀察兩者訊號的功率頻譜密度(PSD)，探討心音與心率變異度(HRV)的關聯性，實現以心音計部分取代心臟超音波之目標。

The traditional auscultation uses stethoscopes to make preliminary diagnosis. Furthermore, doctors utilize the echo-cardiography to evaluate the patients’ cardiac structure. However, there are some shortcomings which include high cost and time on diagnosis. In this thesis, we cooperated with the Children Cardiology doctors in the Kaohsiung Veterans General Hospital and Kaohsiung Chang Gung Memorial Hospital. The thesis implemented the Audicor device and collected 108 clinical heart sound signals and analyzed these existed signals.
These clinical data are classified into normal and abnormal through high-price ultrasound reports. This thesis achieved the fast Fourier transform (FFT) method to transfer the measured PCG and ECG signals from time domain into frequency domain. We implemented band-pass filter (BPF) to find out the differences between the normal and abnormal signals. Then, this thesis simultaneously matched the physiological information on PCG and ECG. In additions, we analyzed the power spectral density (PSD) and discussed the relationship between the heart sound and heart rate variability (HRV). In this thesis, we hope to achieve the goal that the echo-cardiography can be replaced by the the Audicor device.

中文摘要 I
ABSTRACT II
Acknowledgements III
List of Figures 3
List of Tables 5
Abbreviations 6
Chapter 1 Introduction 7
1.1 Motivation 8
1.2 Thesis organization 9
Chapter 2 Background and Related Works 10
2.1 The First and Second Heart Sounds 10
2.1.1 The First Heart Sound and Physical Meanings 11
2.1.2 The Second Heart Sound and Physical Meanings 12
2.2 The Third and Fourth Heart Sounds 14
2.2.1 The Third Heart Sound and Physical Meanings 14
2.2.2 The Fourth Heart Sound and Physical Meanings 15
2.3 Splitting Sounds 16
2.3.1 The First Splitting Sounds 17
2.3.2 The Second Splitting Sounds 18
2.4 Heart Murmurs 21
2.4.1 Innocent Heart Murmurs 22
2.4.2 Pathological Murmurs 24
2.5 Extra Heart Sounds 25
2.6 Introduction of Hardware 27
2.6.1 Cardiac Ultrasound Equipment 27
2.6.2 Audicor RT with AM (Holter) 28
Chapter 3 Methodology of Analyzing the Heart Sounds Signals 30
3.1 The Representation and Analysis in Time-Domain 31
3.2 Fast Fourier Transforms 32
3.3 The Representation and Analysis in Frequency-Domain 36
3.3.1 Band-pass Filter 36
3.3.2 Power Spectral Density 38
3.3.3 Wavelet Transform 40
Chapter 4 Experimental Results and Discussions 42
4.1 Analyzed Results of the First and Second Heart Sound 45
4.1.1 The First Heart Sound 45
4.1.2 The Second Heart Sound 46
4.1.3 The Combination of the First and Second Heart Sound 48
4.2 The Relationship between HRV and Heart Sound 50
Chapter 5 Conclusions and Future Works 53
5.1 Conclusions 53
5.2 Future Works 53
References 54
List of Publication 57
Biography 58

[1] W.H. Tsai, S.H. Su, C.H. Ma, "Unsupervised classification of heart sound recordings", Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2013 Asia-Pacific, Kaohsiung, 29 Oct. -1 Nov. 2013, pp. 1-5.
[2] S. Rajan, E. Budd, M. Stevenson, R. Doraiswami, "Unsupervised and Uncued Segmentation of the Fundamental Heart Sounds in Phonocardiograms Using a Time-Scale Representation", Engineering in Medicine and Biology Society, 28th Annual International Conference of the IEEE, New York, NY, 30 Aug. -3 Sept. 2006, pp. 3732-3735.
[3]Kumar, D. , Carvalho, P. , Antunes, M. ; Henriques, J., Sa e Melo, A., Schmidt, R., Habetha,J.,“Third Heart Sound Detection Using Wavelet Transform-Simplicity Filter”, Engineering in Medicine and Biology Society, 2007. EMBS 2007. 29th Annual International Conference of the IEEE , Page(s): 1277 - 1281.
[4]P. Hult, T. Fjllbrant, K. Hildn, U. Dahlstrm, B. Warnne, P. Ask,”Detection of The Third Heart Sound Using A Tailored Wavelet Approach: Method Verification”, J. of Medical & Biological Engg.& Comp., 43(8), 2005, page. 212–217.
[5]C. Ahlstrom, P. Hult, P. Ask, ”Detection of the 3RD Heart Sound using Recurrence Time Statics”, in Proc. of ICASSP, Touluouse, 2006, page. 1040-1043.
[6]Leung, T.S., White, P.R., Cook, J., Collis, W.B., Brown, E.,Salmon, A.P. ,” Analysis of the second heart sound for diagnosis of paediatric heart disease”, Science, Measurement and Technology, IEE Proceedings, pp. 285-290, August, 6, 2002.
[7]Markaki, M., Germanakis, I., Stylianou, Y. Acoustics, ” Automatic classification of systolic heart murmurs”, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference , Publication Year: 2013 , Page(s): 1301 – 1305.
[8]Yoshida, H., Shino, H., Yana, K.,” Instantaneous frequency analysis of systolic murmur for phonocardiogram”, Engineering in Medicine and Biology Society, 1997. Proceedings of the 19th Annual International Conference of the IEEE , Publication Year: 1997 , Page(s): 1645 - 1647 vol.4.

[9]Shino, H., Yoshida, H., Yana, K., Harada, K., Sudoh, J., Harasewa, E.,” Detection and classification of systolic murmur for phonocardiogram screening”, Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE , Publication Year: 1996 , Page(s): 123 - 124 vol.1.
[10]Noponen, A.-L., Lukkarinen, S., Angerla, A., Sikio, K., Sepponen, R.,” How to recognize the innocent vibratory murmur”, Computers in Cardiology 2000, Publication Year: 2000 , Page(s): 561 – 564.
[11]www. heart.org /HEARTORG/Conditions/More/HeartValveProblemsandDisease/ When-A-Heart-Murmur-Signals-Valve-Disease_UCM_450616_Article.jsp
[12]Shino, H., Yoshida, H., Yana, K., Harada, K., Sudoh, J., Harasewa, E.,” Detection and classification of systolic murmur forphonocardiogram screening”, Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE , Publication Year: 1996 , Page(s): 123 - 124 vol.1.
[13]Taikang Ning, Kai-Sheng Hsieh., “Automatic heart sounds detection and systolic murmurcharacterization using wavelet transform and AR modeling”, Engineering in Medicine and Biology Society (EMBC), 2013 35th Annual International Conference of the IEEE, Publication Year: 2013 , Page(s): 2555 – 2558.
[14]Sparto, P.J., Parnianpour, M., Barria, E.A., Jagadeesh, J.M., “Wavelet and short-time Fourier transform analysis of electromyography for detection of back muscle fatigue”, IEEE Communications Magazine, vol. 43, no. 3, pp. 433-436, Sep. 2000.
[15]Vikhe, P.S.; Nehe, N.S.; Thool, V.R., “Heart Sound Abnormality Detection Using Short Time Fourier Transform and Continuous Wavelet Transform”, Emerging Trends in Engineering and Technology (ICETET), 2009 2nd International Conference, IEEE, Publication Year: 2009 , Page(s): 50-54.
[16]Gavrovska, A.M., Slavkovic, M.S., Paskas, M.P., Dujkovic, D.M.,Reljin, I.S., “Joint time-frequency analysis of phonocardiograms”, Neural Network Applications in Electrical Engineering (NEUREL), Publication Year: 2012 , Page(s): 177 – 180.

[17]Arvin, F., Doraismay, S., “Real-time segmentation of heart sound pattern with amplitude reconstruction”, Biomedical Engineering and Sciences (IECBES), 2010 IEEE EMBS Conference, Publication Year: 2010 , Page(s): 130 – 133.
[18] Vikhe, P.S., Hamde, S.T., Nehe, N.S., “Wavelet Transform Based Abnormality Analysis of Heart Sound”, Advances in Computing, Control, & Telecommunication Technologies, 2009. International Conference, IEEE , Publication Year: 2009 , Page(s): 367-371.
[19]心臟功能室檢查儀器介紹，No26第4卷第3期2008-05出刊。http://www.edah.org.tw/journal/26/51-56.html
[20]AUDICOR RT with AM (Holter) Quick Reference Guide
[21]AUDICOR RT with AM (Holter) Quick Reference Guide, Addendum
[22]http://www.cs.princeton.edu/~fiebrink/314/2010/week8/FFT_handout_2010.pdf
[23]超音波基本原理及儀器簡介。http://yct2004.pixnet.net/blog/post/38124247
[24]http://www.stethographics.com/heart/main/sites.htm
[25]http://www.learntheheart.com/cardiology-review/s1-heart-sound/
[26]http://file.scirp.org/Html/9-20469_30997.htm
[27]Simon Haykin, textbook of “Communication Systems”, 4th edition.
[28]Wikipedia, http://en.wikipedia.org/wiki/Spectral_density
[29]Wang, X.L. ; Eklund, J.M. ; McGregor, C. , “Parametric Power Spectrum Analysis of ECG Signals for Obstructive Sleep Apnoea Classification”, Computer-Based Medical Systems (CBMS), 2014 IEEE 27th , Publication Year: 2014 , Page(s): 8 – 13.
[30]http://whatis.techtarget.com/definition/bandpass-filter
[31]Taiwan Wiki, http://www.twwiki.com/wiki/
[32]http://www.easyauscultation.com/
[33]George E. Billman, “The LF/HF ratio does not accurately measure cardiac sympatho-vagal balance”, frontiers in Physiology, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3576706/
[34]HRV與身心健康。心律大師®/ ANSWatch® 官方網站。http://taiwanscientific.com.tw/index.html