Research And Application Of A Heart Sound Generation Model Based On Cascaded Lossless Acoustic Tube

During the cardiac cycle,because of systolic and diastolic,a series of mechanical vibration is caused by valve opening and closing,the blood flow impacting the ventricular wall and artery and other factors,which produces the heart sound.The heart sound contains the relationship between the internal structure of the heart and the sound process,and the characteristics of the heart sound have some internal relations with the internal shape of the heart.Therefore,the main purpose of the research on heart sound model is that we put forward a heart sound model,which can help us to further understand the mechanism of heart sound;by using the model on the study of abnormal heart sounds,it can provides a reliable theoretical basis for the diagnosis analysis of abnormal heart sounds.In this paper,a heart sound generation model based on cascaded and nondestructive acoustic tubes is presented.Besides,the model is applied to the classification recognition and analysis of abnormal heart sounds.First of all,based on the acoustic theory,combined with the traveling wave theory,this paper presents a heart sound generation model based on cascaded and nondestructive acoustic tubes,which consists of a S1 cascaded and nondestructive acoustic tubes model and a S2 cascaded and nondestructive acoustic tubes model.In the model,the ventricles and arterial inside of the heart are abstracted into multistage tubes with the equal length and the different radius.By controlling the radius of the acoustic tubes,the process of relaxation and constriction of the ventricles and arterial is simulated.Then,the formant frequency is used to simulate the process of heart sound generation.This model has the following three characteristics.First,as heart sound is an acoustic signal,this model uses the formant frequency as the target parameter,which can directly reflects the acoustic characteristics of the heart sound.Second,the ventricles and arterial are abstracted into multistage tubes with the equal length and the different radius,which makes the relations between the model and the shape of heart,highlights the relation between heart sound model and the generating mechanism of heart sound.So,this model has a positive effect on the exploration of the heart sound generation principle.Third,by using 9 inner radius {Rm}(m,1,2,3...9)as parameters of the model,the model is simple and intuitive.And the parameters of model are easy to control.Besides,as the target parameters,formant frequency of S1 and S2 has the advantages with low vector dimension and clear physical meaning.Secondly,the abnormal heart sounds are simulated through the model.By comparing with theexperimental results of the normal ones,the abnormal heart sounds are analyzed pathologically,the analysis results are consistent with the clinical manifestations.Finally,based on formant theory,a sound recognition method with the characteristic of S1 and S2 is put forward.As the feature vector,formant frequency can characterize the energy distribution of heart sound well.And it has a high recognition rate when formant frequency is applied on recognition of heart sound.Through the model,both normal and abnormal heart sound be simulated very well.From the experimental results,it shows that the model has the positive significance for understanding the heart sound production process and analysing the causes of the pathology of heart sounds.