The Design And Realization Of Algorithm For Identification Of A-V Junctional Escape Beat And Atrial Fibrillation&Atrial Fibrillation Based On Single-lead ECG Signal

Electrocardiogram(ECG),as a reliable and clear information carrier that reflects the state of the heart,plays an important role in the diagnosis of heart disease.The development and application of ECG automatic diagnosis technology can firstly improve the accuracy and efficiency of electrocardiogram analysis to a large extent,so as to assist in accurate diagnosis of heart disease.Secondly,it can greatly reduce the burden on ECG doctors and thirdly provide objective evidence for clinical diagnosis and avoid the dependence of diagnosis results on the subjectivity and experience of clinicians.At the present stage,Nodal(junctional)escape beat classification and identification is mainly based on learning algorithms.The existing methods require large amount of data and is time consuming.These shortcomings limit ECG-based Nodal(junctional)escape beat real-time analysis.Although atrial fibrillation and atrial flutter recognition have achieved a good classification and recognition effects in many methods,the rigidity of these methods for different ECG acquisition apparatus is also required to be improved.For this reason,this paper proposes a Nodal(junctional)escape beat algorithm based on waveform features,which can realize efficient and accurate identification of the Nodal(junctional)escape beat appearing in 1-minute ECG data,and can meet the real-time requirements.Then,this paper proposes the identification algorithm for atrial flutter and atrial fibrillation based on higher-order statistics.The specific research contents are as follows:(1)Design and implement an algorithm identifying Nodal(junctional)escape beat based on the RR interval and characteristic P wave.We divide the ECG signal into segment of 6 second per segment.And input segment by segment.Firstly,we find the position of P wave,Q wave,R wave and S wave.Secondly,calculate the PR interval,RR interval,the correlation coefficient of two adjacent QRS complex waves,and the state of P wave(standing,absent or inverted).The three are taken as features and combined to identify the Nodal(junctional)escape beat.This algorithm was verified using real ECG data from the MIT-BIH arrhythmia database.Experimental results show that the algorithm can achieve a recognition accuracy of 93.7%for the Nodal(junctional)escape beat while ensuring high efficiency.(2)Design and implement identification algorithm of atrial flutter and atrial fibrillation based on higher-order statistics.Using the significant differences among the SQ segments of atrial flutter,atrial fibrillation,and sinus rhythm,we extracted those data segments with difference to calculate the high-order statistics.The atrial flutter atrial fibrillation are classified and identified by KNN classifier using the extracted quantitative features.Using the real ECG data from MIT-BIH arrhythmia database to verify the algorithm and compare it with related algorithm.The results show that the algorithm can achieve better classification than other related algorithm in identifying atrial flutter and atrial fibrillation.