Research On The Method Of Coronary Heart Disease Recognition Based On Pulse Transit Time Variability

Coronary heart disease is one of the greatest hidden dangers to human health with its high incidence,low cure rate and other characteristics.If there is no effective prevention and treatment,coronary heart disease will become a serious problem for the development of mankind in the future.With the rapid development of intelligent medical equipment,it was brought convenience to the diagnosis and treatment of coronary heart disease.However,coronary heart disease recognition also has low accuracy,long time-consuming detection and costly and other defects.Therefore,It is very important to research a high real-time and accurate method of coronary heart disease.Electrocardiogram(ECG)signal and pulse signal contains abundant physiological and pathological information that can be used as indicators of prevention and recognition of coronary heart disease.The ECG signal and pulse signal is combined to detect the pulse transit time variability(PTTV)signal,the PTTV signal can reflect the severity of coronary artery disease and the autonomic nervous system regulation mechanism.Real-time analysis of PTTV signal and extraction of relevant information has important significance for real-time monitoring and early warning of coronary heart disease.Based on the pathologic mechanism and clinical diagnosis of coronary heart disease,this thesis summarized the clinical diagnosis research and the recognition research status of coronary heart disease.On the basis of the review method,by means of the severity of coronary artery disease and autonomic regulation principle in the course of coronary artery disease,PTTV was proposed to be used to achieve real-time and accurate recognition of coronary heart disease.The methods in this thesis solved the problem in the existing research that the recognition of coronary heart disease is achieved only by the autonomic nervous system regulation recognizes and that the real time and accuracy of the extraction of PTTV is cared for this and lost that.The main work in this thesis is as follows.1)The extraction method of PTTV signal was researched.The time interval of R wave of synergistic ECG signal to peak of the main wav e of synergistic pulse signal was PTTV,it was extracted by comparison of the data features and the existing analysis methods.The high real-time integer coefficient filter was used to the variety of noise and interference of ECG and pulse signal in the collection process.2)For the situation that the analysis methods of PTTV signal are time-consuming,subjective and so on,the methods of time domain and non-linear analysis were improved by the idea of sliding window iteration based on the characteristics of PTTV signal.The experiment verified that the improved feature extraction methods are real-time and accurate.Meanwhile,in this thesis,it was found that the distribution of spectral energy of PTTV signal is more obvious than the distribution of spectral energy of ECG signal.3)According to the characteristics of PPTV signal,the model parameters of each recognition algorithm were determined,and the importance of parameters was further explained by recognition accuracy and algorithm running time.Meanwhile,the t-test and principal component analysis were used to select the feature,which effectively retained the original feature information and reduced the data dimension,thus the complexity of recognition algorithm was reduced.Through the comparative analysis of the experiment,a real-time and accurate algorithm was proposed to be used to coronary heart disease recognition.