| More than 300,000 children with congenital heart disease(CHD)are born in China every year,high incidence and low screening rate of CHD are widespread especially in western mountain plateau areas with a poor medical condition,CHD threat to children's growth and development seriously.Due to phonocardiogram(PCG)is closely related to cardiac structure and function,many studies have focused on the identification of heart disease through PCG signal processing,and made some progress.However,there are currently few studies on intelligent diagnosis of pediatric murmurs due to congenital heart disease,and the mechnism of heart sound has been ignored.The research background of this paper is intelligent auscultation for screening of children with congenital heart disease.Firstly,a method of intelligent diagnosis of pediatric CHD murmurs was developed.Then the selectivity problem of cardiac auscultatory areas under electronic stethoscope was studied.Finally,the fluid mechanics mechanism of heart sound was preliminarily explored.The main research contents are as follows:1.In order to develop an intelligent diagnosis method for heart murmurs in children with congenital heart disease.Phonocardiogram(PCG)signals of 86 children were recorded with 24 children in normal heart sounds and 62 children in CHD murmurs.A segmentation method based on the discrete wavelet transform combined with Hadamard product was implemented to locate the first and the second heart sounds from the PCG signal.Ten features specific to CHD murmurs were extracted as the input of classifier after segmentation.Eighty-six artificial neural network classifiers were composed into a classification system to identify CHD murmurs.In order to realize the visualization of intelligent diagnosis process,graphical user interface of PCG intelligent auscultation system was developed based on Matlab.The accuracy,sensitivity and specificity of diagnosis for heart murmurs was 93%,93.5%and 91.7%,respectively.In conclusion,a method of intelligent diagnosis of pediatricCHD murmurs is developed successfully,the diagnosis method is convenient and efficient,and can be used for on-line screening of CHD in children.2.In order to explore the correlation between the selectivity of cardiac auscultatory areas and heart murmur detection under electronic stethoscope,comparison of PCG signals under five traditional auscultatory areas of the same patient with CHD was performed.Firstly,the time-frequency distribution was compared.Secondly,heart sound signal quality detection system was designed for quality evaluation.Finally,the five groups of PCG signals were input into the PCG intelligent auscultation system designed in this paper for murmur detection.The comparison result showed that the PCG signal at the best auscultatory area had the highest quality and the most high-frequency murmur components,but heart murmurs were both detected at five traditional auscultatory areas.In conclusion,the selection of cardiac auscultatory areas is not related to heart murmur detection under electronic stethoscope.Although the qualities of five groups of PCG signals are different,all heart murmurs can be detected when placed electronic stethoscope on the approximate position of the heart surface.This study provides theoretical support for the non-professional recording of PCG signals in intelligent auscultation.3.In order to explore the fluid mechanics mechanism of heart sound preliminarily,the simulation experiment platform of heart valve jet was built based on the left heart of isolated pig heart.Jet sound signal of different leakage orifices of the mitral valve in static and dynamic of 120 mmHg was recorded.The time-frequency distribution and features of the jet sound signal were analysed.A mathematical model for predicting the area of valve leakage orifice based on the jet sound signal was proposed.The experimental results were as follows: 1)The mostly frequency of the jet sound signal was ranged from 0Hz to 200 Hz,a little ranged from 200 Hz to 400 Hz.2)In static and dynamic experiments,with the increase of the leak orifices,the time-frequency distribution and features of the jet sound signal showd trending changes.3)The prediction relative errors of valve leakage orifice area of different jetstates were both within 5%.The prediction model is accurate and reliable,and the experiment provides basic research for the mechanism of cardiac jet sound. |
PDF Full Text Request