Abstract
ECG analysis diagnosis system mainly consists of two phases: waveform recognition and intelligent diagnostics. In practical applications, the ECG waveform recognition is the key to the system. Accuracy and reliability of the detection determine the diagnosis and treatment of heart disease. There are a variety of detection ways mentioned in the relevant papers. In this paper, a waveform recognition algorithm based on geometric characteristics is proposed. At first the ECG signal is filtered by the digital filter such as Butterworth filter, mean filter and so on in this proposed method, and then its first derivative and second derivative are computed respectively, which we employ to determine the trend and to compute the optimum slope of each point. We recognize the peak, onset, offset of ECG waves combined with the geometric features. At last by computer simulation with the PTB diagnostic ECG database, the algorithm is feasible.