Dissertation > Excellent graduate degree dissertation topics show

The Analyses and Researches of Heart Sound Signal on the Basis of Matlab

Author: YangYanNi
Tutor: YanBiGe
School: Shaanxi Normal University
Course: Acoustics
Keywords: Heart sound signals Wavelet Transform STFT Autoregressive model
CLC: R318.04
Type: Master's thesis
Year: 2007
Downloads: 483
Quote: 4
Read: Download Dissertation

Executive Summary

Showed an increasing trend with modern social and material living standards improve, cardiovascular disease, its cause mortality among the first of the various diseases, as the biggest threat to human health. The heart sounds is a reflection of the condition of the mechanical movement of the heart and cardiovascular system, which includes the role of the various parts of the heart itself, and between physiological and pathological information. Analysis of heart sounds with modern digital signal processing technology has significant value in basic research and clinical diagnosis of cardiovascular disease. Heart sound research field of the study of the physiology and pathology of the first heart sound and second heart sounds; non-invasive detection of artificial heart valves; analysis of heart sounds weak ingredients (third heart sound and fourth heart sounds); analysis of heart murmur frequency variation; positioning extract from a cardiac cycle of heart sounds ingredients; conduction mechanism modeling of the heart sounds. The basis for discussions on the heart sound signal analysis method, pre-processing and ingredients positioning, extract the characteristic frequency of the second heart sound different pathological cases, provide a reference for the clinical diagnosis of heart disease. According to the valvular theory, the second heart sound (S 2 ) from ventricular diastolic ventricular wall vibration, the aortic valve (Aortic valve) and pulmonary valve (Pulmonary valve) off and atrioventricular valve open flow from the atrium into the ventricle, comprising a plurality of frequency components is an important part of the heart sounds. In this paper, the short-time Fourier transform and AR model power spectrum analysis and extraction of the second heart sound data of 20 cases of heart sounds (including normal 10 cases, 10 cases of abnormal) two main characteristic frequency, combined with the physiological characteristics of the experimental results of the second heart sound were analyzed and discussed. The experimental results show that: (1) the second heart sound (S 2 ) containing aortic lobe (A 2 ) and pulmonary valve (P 2 ) two main ingredients; (2) (A 2 ) of abnormal heart sounds (S 2 ) aortic and pulmonary valve (P 2 ) frequency than normal heart sounds increased; (3) the second heart sound also contains low-frequency, low-amplitude, longer duration ingredients. Mainly in the following aspects: 1. Heart sound data collected from the hardware to the very serious interference, the heart sound signal can not be normal analytical processing, these disturbances include the 50Hz frequency interference, the interference of breath sounds and white noise. According to the different signal and noise wavelet coefficients after wavelet transform with the scale changes, the paper proposes denoising using wavelet multi-resolution analysis of heart sound signals. Through comparative analysis, to find the best wavelet denoising parameters, ie coif5 Wavelet heart sound signal layer decomposition de-noising by soft thresholding and fixed threshold. 2. Analysis of heart sounds, time-varying characteristics of heart sound signals, using the following method and compare their advantages and disadvantages in the analysis of heart sound signals, different analytical methods in order to be selected depending on the purpose. (1) heart sounds, heart murmurs frequency components is an important feature of the heart sounds, therefore, the use of modern power spectrum AR model analysis method of heart sounds on a cycle energy analysis, divided into high, medium and low frequency in the frequency component of the heart sounds bands. (2) heart sounds is a time-varying signal its time - frequency characteristics that best reflect the characteristics 2205Hz sampling frequency of 128-point Hamming window short-time Fourier transform of heart sound signals, the results of the analysis can be used two-dimensional and the three-dimensional two ways to display the results intuitive. (3) Another method is to time-frequency analysis of the heart sound analysis, The dyadic wavelet sub-band filter characteristics, the heart sound is decomposed into three bands of high, medium and low frequencies, the use of the signal waveform of each band reconstruction Figure examine each band of the time - amplitude characteristics of each ingredient and its trends, and characteristics of the heart sounds made fully reflected. 3. Using wavelet multiresolution analysis method and normalized the Shannon energy extraction second heart sound. Multiresolution analysis by wavelet after decomposition of the heart sound signals, different frequency bands are reconstructed, and then each band is the same length of time segment, computing for each segment the Shannon normalized energy, and then, combined heart sound characteristic determination S < sub> 2 of the location. 4. Eigenvalue extraction of the second heart sound signal. AR model and the short-time Fourier transform to detect the two main characteristics of the frequency of the second heart sound, comparative study of the different points of the normal heart sounds and heart sounds valvular heart disease, and provide a reference for the clinical diagnosis of heart disease. This article provides a basic analysis methods clinical studies of the heart sound signal, laid the foundation for the heart sounds for assisted diagnosis of heart disease.

Full-text Catalog

Abstract     3-5
Abstract     5-10
first chapter     10-13
1.1 of this topic significance     10
1.2 This topic research status     10-12
1.3 this paper the main content and the work done     12-13
Chapter II heart sound signals of Medicine explore     13-17
2.1 heart sounds to form the physiological mechanisms     13-14
2.2 heart sound time-domain characteristics     14-15
2.3 heart sound signal frequency characteristics     15
2.4 heart sounds with valvular heart disease relationship     15-17
third heart sound signal analysis method comparison studies     17-31
3.1 heart sound signals used in the analysis environment     17
3.2 heart sounds of the AR model spectrum estimation     17-20
3.3 short-time Fourier transform     20-21
3.4 wavelet transform     21-28
3.4.1 continuous wavelet transform     23
3.4.2 discrete wavelet transform     23
3.4.3 binary wavelet transform     23-24
3.4.4 wavelet transform multi-resolution analysis     24-25
3.4.5 Mallat algorithm - fast dyadic wavelet decomposition and reconstruction algorithm     25-27
3.4.6 heart sound signal wavelet transform     27-28
3.5 analysis method of discussion     28-30
3.6 Summary     30-31
Chapter heart sound signal wavelet denoising     31-42
4.1 Introduction     31
4.2 noise the wavelet decomposition characteristics     31-32
4.3 wavelet noise reduction methods and steps     32-33
4.4 mother wavelet function select     33-36
4.4.1 Haar wavelet     33-34
4.4.2 Daubechies wavelet coefficients     34-35
4.4.3 Symlet wavelet coefficients     35-36
4.4.4 Coiflet wavelet coefficients     36
4.5 wavelet decomposition layers     36
4.6 the wavelet denoising closing value of selection rules     36-38
4.7 computer simulation analysis     38-41
4.8 Summary     41-42
Chapter qualitative and quantitative comparative analysis of the abnormal heart sound signal     42-49
5.1 experimental data, select     42
5.2 extract the second heart sound     42-46
5.2.1 the heart sounds ingredient extracted     42-43
5.2.2 extracted based on the wavelet transform of heart sound components     43-46
5.3 positive abnormal heart sound signal analysis and feature extraction of     46-48
5.4 Summary     48-49
Chapter 6 Conclusions and Prospects     49-51
6.1 summarizes     49-50
6.2 Job Outlook     50-51
References     51-54
Acknowledgements     54-55
to research studying for a degree during     55

Similar to the Dissertation

  1. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  2. Research on Identification System of Cashmere and Wool Fiber,TS101.921
  3. Cloning of StFT and StAP1 from Solanum Tuberosum and Analysis of Tuberization in Constans-Antisense Transgenic Potato Plants,S532
  4. Characteristics of sensory stimulation evoked,R318.0
  5. Network transmission ROI image coding algorithm,TN919.81
  6. Image Fusion Algorithms Based on Multi-scale Analysis,TP391.41
  7. Feature Extraction Technologies Research and Implementation of 3D Models Based on Wavelet Transform,TP391.41
  8. An Algorithm on Clustering and Anomaly Detection for Multiple Data Streams,TP311.13
  9. Research on Information Hiding Based on Images,TP309.7
  10. Design and Implication of Image Compression System Based on Wavelet and DM6446,TP391.41
  11. Combined DWT Dynamic Data Reconciliation Research and Application,TP274
  12. Traffic Prediction Model Based on Wavelet and Markov,TP393.06
  13. The Study of Image Watermark Base on Discrete Wavelet Transformation,TP309.7
  14. Applied Research of Underdetermined Blind Source Separation Method on Phonocardiogram Mixed Signal,TN911.7
  15. The Compression and Application of the Brain CT Image,TP391.41
  16. Inspection and Recognition on a Class of Tool Processing Trace,TP391.41
  17. The Research and Design of a Fire Detection System Based on Combustion Audio Recognition,X924.4
  18. Surface EMG Signal Analysis and Real-time System Construction,R318.0
  19. The Realization and Study of a Scalable Data Encoder Based on SPIHT,TN762
  20. Research and Implementation of Dwt-based Video Codec System at Low Bit Rates,TN919.81
  21. Improved ROI Coding Method Based on SPIHT,TN919.81

CLC: > Medicine, health > Basic Medical > Medical science in general > Biomedical Engineering > General issues > Biological information, biological control
© 2012 www.DissertationTopic.Net