Research On Algorithm And Embedded Software Of Triad ECG Detection System

Clinical ECG detection is divided into static ECG, dynamic ECG, exercise ECG, emergency ECG, of which the first three have become conventional methods of clinical diagnosis. In order to realize the different requirements of sampling rate, sample precision, bandwidth, power and volume of ECG signal, the current clinical ECG instruments are special instrument, which are difficult to meet the needs of the community, first aid, aerospace and other fields to cover for multiple types of heart disease diagnosis. Therefore, it is of great significance to develop a kind of instrument which has the function of static ECG, dynamic ECG and exercise ECG detection.This thesis studies and designs a three-in-one ECG detection system, on the premise of hardware guarantee sampling rate and accuracy, a scheme of different ECG detection is proposed, which is based on embedded software and algorithms, thus realized static ECG, dynamic ECG and exercise ECG test in a hardware.For guarantee the minimum baseline drift of exercise ECG, and guarantee the characteristics of T wave and QRS wave form, the paper proposed an algorithm of ECG signal filtering based on empirical mode analysis. Adaptive decomposition of the original signal, through the comparison and analysis of the correlation between scales and ECG features, select part scale for signal reconstruction, threrby in the removal of interference while better the features of the ECG retain. Implement embedded software design based on Cortex-M3 processor, including ECG data acquisition, storage, communication and multi task management, data flow control, as well as static ECG, dynamic ECG, exercise ECG filtering. The low power consumption design and reliability design are carried out, which can meet the requirements of the safety of medical instrument.System test results show that the data processing is reliable, two five battery can support 36 hours of continuous work, the system can pass static test and quickly restore the fault before state of work, functional tests verify that the system can achieve three ECG. Clinical trials showed that the system has a good waveform detection of ECG, baseline drift is small and so on. In clinical practice, the data collected by our system and the motion plate (GE Marquardt CASE motor ECG) are in good agreement.The system studied in this paper can meet the requirements of high precision detection of three kinds of ECG signals, which has important significance for the clinical diagnosis of cardiovascular disease and the special field of ECG monitoring.