Design And Implementation Of ECG Analysis System

In recent years,with the more and more phenomenon of sudden death among young people,the health problems of the heart have been talked about and pa id more attention again,which lead to the sudden prosperity of wearable devices.In fact,heart diseases are paid special attention to the theme of the medical field over the years,which have a great variety,high secretiveness,high mortality,etc.,and diseases threat greatly on human health and life safety.Therefore,the study of ECG signal is very necessary and great of significance,including ECG preprocessing,ECG features detection and diagnosis of arrhythmia by ECG features detection acquisition.The results of ECG preprocessing determine the accuracy of ECG waveform features detection,and the ECG features detection results directly affect the classification and diagnosis on related diseases of arrhythmia.In view of the above problems,this paper mainly focuses on the three aspects on ECG signal preprocessing,ECG waveform features detection and classification of arrhythmia diseases.First of all,designs and implementations of data reading algorithm according to the storage rules of ECG signal data files were proposed.Then according to the characteristics of ECG signal noise,multi-scale wavelet decomposition and reconstruction methods to design ECG signal denoising algorithm and effects with the traditional filter algorithm for comparative analysis were given.Secondly,the ECG waveform recognition algorithms of the denoised ECG signals were proposed.A nonlinear amplification R wave detection algorithm was designed and implemented according to the features of the ECG waveform.In addition,according to the location characteristics of ECG waveform,the design and implementation of a new fast R wave detection algorithm was proposed,with construction of pulse wave by location characteristics of ECG waves.The corresponding algorithm of missing detection and error detection were proposed.What’s more,the results of R detection were verified and the correct ratio of R wave detection was given.The ideas of construction of pulse wave and location on Q wave,S wave and their start and end positions of points were proposed.Design and implementation of P wave,T wave,and their start and end positions of points detection algorithm were given.The detection results of each ECG waveform were analyzed.Finally,using the ECG waveform detection results to extract ECG parameters and create decision rules based on clinical diagnosis knowledge.And using the rules to create constructing decision tree algorithm to get classification results and diagnosis of several arrhythmia diseases.And the results were analyzed.On the basis of the above three aspects,designs and implements the ECG analysis system using the Qt software platform under Windows were proposed in this paper.Through the operations and results of the software show,this software runs smoothly without errors and has a good effect on ECG signal pretreatment.This software shows ECG waveform correct detection results,and several kinds of arrhythmia diagnosis have a good effect on ECG automatic classification.This work shows the study has a reference value for analysis of workers on ECG.