Research On The Relations Between The Features Of First Heart Sound And Cardiac Hemodynamics

Heart sounds are produced by the mechanical vibrations of cardiovascular tissues which including heart wall, heart valves and arteries. The mechanical vibrations are caused by the cardiac hemodynamics. When the cardiac hemodynamic changes, the features of heart sound will produce corresponding changes. In this paper, relations between the amplitude and timing features of first heart sound (S1) to cardiac hemodynamic are investigated, and mathematical models for the quantitative description of the above relations are given. Then, the features of S1can be used to predict cardiac hemodynamic according to the mathematical model which can provide a technological approach to monitor or evaluate cardiac function. In this paper, the main work is focused on the following two aspects.(1) The relation between the amplitude of S1and the rising rate of left ventricular pressure is investigated. In animal experiment, epinephrine was used to evoke the dog’s cardiac hemodynamic change. After injecting different doses of epinephrine into the dog, the electrocardiogram, external phonocardiogram and left ventricular pressure signals were recorded synchronously. The amplitudes of the S1and rising rates of left ventricular pressure are obtained after preprocessing the collected signals. According to the standardized residual and leverage ratio, outliers can be removed. In order to investigate the relation between the maximum amplitude of S1and the maximum rising rate of left ventricular pressure quantitatively, the linear, quadratic, cubic, and exponential models are proposed to fit the relation between them. According to the goodness of fit index and the extension of models, the relation between them fits the exponential model better. Furthermore, the amplitudes and rising rates of left ventricular pressure of the first two peaks or valleys of S1are obtained, and the results of data analysis show that relations between them are still fit the exponential model.(2) The relation between the timing delay from the R-wave to maximum peak of S1(TDRMP) and respiratory phase (RP) is investigated. The previous studies found that respiration can change cardiac hemodynamic and lead to a delay or earlier occurrence of the S1. Electrocardiogram, external phonocardiogram and respiratory signals were recorded synchronously under spontaneous respiration. It is observed that the relation between TDRMP and RP is nonlinear, i.e. TDRMP decreases with expiration and increases with inspiration. Hammerstein-Wiener model is proposed to investigate the nonlinear relation between TDRMP and RP. The model’s input-output equation is simplified by the key term separation principle. An iterative algorithm is used to estimate the model’s parameters. Based on the above method, data from12experiments are analyzed. The results show that there is a better fit between the predictions and the observations of TDRMP. The mean of goodness of fit index and MSE are94.56%and1.11%, respectively. This study indicates that the Hammerstein-Wiener model can describe the nonlinear relation between TDRMP and RP effectively in time domain.