Research On Dynamic Electrocardiogram Signal Quality Evaluation And Disturbance Analysis

Heart disease is a common disease of the circulatory system,a serious threat to the health of human.The electrocardiogram(ECG)is an effective means of heart disease detection in modern medical field.Because of the acquisition of ECG signal is highly susceptible to the influence of some interference factors,this thesis has done in-depth researches on the algorithms for assess the quality of ECG signals and automatic identification of motion artifact in the Holter system.Through the evaluation of the quality of ECG signal and identification of motion artifact,it can effectively improve the accuracy of diagnosis,which has important scientific research value and urgent realistic demand.According to the main interference sources affecting the ECG signal and the physiological significances of ECG signal frequency range were analyzed,the signal ratio,signal energy entropy in the six different bands were extracted as the ECG signal quality indexs.Then the quality of ECG signal in the database was assessed based on the signal quality indexs.The quality of ECG signal in the database was assessed respectively using support vector machine(SVM)classification method.The method of support vector machine(SVM)classification,was about to select the best parameters according to the ECG signal quality index in the training set,train the support vector machine according to the selected parameters.Then using the trained support vector machine to assess the quality of the ECG signals in the test set,which was collected in the Holter system.By analyzing the QRS complex shape variation law of the beats,proposed the algorithms for automatic identification of motion artifact based on beat template.According to the similarity between each beat and the beat template,the beats shape matching curve was calculated.In the curve,fuzzy-logic criterion and local integration were employed to find out the start and end position of the motion artifact segments,achieving the automatic identification of motion interference.The experimental results were compared between this method and other methods,then analyzing the respective advantages and disadvantages.The proposed method in this thesis has been used for many DCG diagnosis applications for greatly improving the analysis of diagnostic algorithms in the dynamic environment.