Research On R-wave Detection Algorithm Of ECG And Its Application

As an important basis for clinical diagnosis of cardiovascular disease,ECG is very important for its accurate automatic detection and analysis.R wave is the most important characteristic main wave in ECG signal.Its accurate location is the premise and foundation of ECG automatic analysis technology.However,when R-wave detection algorithm is applied to mobile ECG equipment,it will be faced with some complex problems,such as strong interferences,individual differences and timeliness requirements.Therefore,in order to improve the accuracy,fast real-time performance,anti-interference and abnormal waveform recognition ability of R-wave detection algorithm,this paper studies different R-wave automatic detection algorithms and their adaptability in mobile devices.(1)Optimization direction and evaluation of R-wave detection: by analyzing the morphological differences of ECG in different acquisition methods,noise environment and pathological waveform,this paper analyzes the influencing factors of R-wave detection and the R-wave detection steps of mobile devices,discusses the key problems,and puts forward a comprehensive evaluation of the algorithm based on the universal evaluation index of R-wave detection and the detection requirements other mobile devices assessments.(2)R-wave detection based on manual feature extraction of digital signal processing: A modified Pan-Tompkins algorithm based on synchronous compression wavelet transform(SSWT)is designed.With the advantages of high resolution and anti aliasing of sswt,combined with the advantages of fast real-time of classical threshold detection algorithm,the R-waves are effectively enhanced,fast and accurate detection is realized,the false and missing detection rates of R-wave are reduced.(3)R wave detection based on self-learning and automatic extraction of key features: a method of R wave automatic location based on convolutional neural network(CNN)is studied.On the basis of the traditional feature classification and fitting method,the design of location steps is further increased,and the input processing and structure of the network are optimized.The R wave can be located quickly and accurately,and the noise and abnormality can be detected Waveform has certain adaptability;(4)Research on the applicability of R-wave detection algorithm in mobile ECG devices: through the comprehensive test of MIT-BIH database and mobile device data set,combined with the detection requirements of mobile devices,the proposed algorithm is evaluated.The results show that sswt-pt algorithm has the advantages of high accuracy and strong interference resistance of single lead detection;CNN algorithm has good recognition ability and calculation cost in special form Good performance: the sswt-pt algorithm has better detection performance in the case of strong noise interference,pathological waveform,or low-quality data.