Extracting Cardiac Dynamics For Cardiac Function Monitoring In Exercise Training

At present,myocardial ischemic diseases pose a major threat to human health,and its incidences is increasing,and there is a development trend to younger.Therefore,it is of great significance for the early diagnosis of myocardial ischemia.Clinically,the ST-T segment of the electrocardiogram(ECG)is generally used to judge the condition of myocardial ischemia.This method has high requirements for the level of a medical personnel.It is only through the naked eye that some information of ECG cannot be seen and has the risk of misdiagnosis.Although coronary angiography is currently available for more in-depth diagnosis of myocardial ischemia,the technique is traumatic to the human body and is expensive and complicated to diagnose.Therefore,it is of great significance to use the engineering technology to make more excavation of the characteristic information in the ECG.The Cardiodynamicsgram(CDG)based on the Deterministic Learning Theory is an electrocardiogram holographic signal obtained by modeling the ECG signal through a RBF neural network.Deterministic Learning Theory modeled the ECG signal using a radial basis neural network and generated a three-dimensional visualization dynamics diagram along the trajectory of the ST-T segment of the ECG.CDG has been initially clinically proven in the diagnosis of myocardial ischemia,providing a new engineering method for early diagnosis of myocardial ischemia.Based on the application of CDG in the diagnosis of myocardial ischemia,in this paper,firstly,we extracted dynamic feature of CDG and designed system for CDG calculation.Secondly,we explored and studied the effectiveness of CDG's dynamic feature in monitoring of exercise training load.This paper focuses on the implementation of CDG dynamic feature extraction algorithm and the design of CDG system.Deterministic Learning Theory and Spatiotemporal Lempel-Ziv complexity algorithm are used as the theoretical to extract the dynamic feature of CDG.We extracts the time dispersion,spatial dispersion,time complexity and spatial complexity of CDG and implemented the feature extraction algorithm based on the C++ programming language.Based on the MFC programming framework,we designed and implemented the CDG system that is simple to use and fast in calculation.This system laid the engineering foundation for the practical application of CDG.Finally,through experiment,the effectiveness of various dynamic characteristics of electrocardiogram in exercise training load monitoring was studied,and preliminary conclusions were obtained.