TY - JOUR AU - Dubrovin, V. I. AU - Tverdohleb, J. V. AU - Kharchenko, V. V. PY - 2014/03/31 Y2 - 2024/03/28 TI - AUTOMATED SYSTEM FOR THE ANALYSIS AND INTERPRETATION OF ECG JF - Radio Electronics, Computer Science, Control JA - RIC VL - 0 IS - 1 SE - Progressive information technologies DO - 10.15588/1607-3274-2014-1-22 UR - http://ric.zntu.edu.ua/article/view/27290 SP - AB - An existing ECG analysis and interpretation software is reviewed in the paper. A wavelet-based ECG delineation algorithm which performs QRS detection and provides as well the locations of the peak(s) of P, Q, R, S, and T waves, and the P, QRS, and T wave boundaries using a single analysis stage: the dyadic wavelet transform of the ECG signal is proposed. A neural network for the classification of heartbeats is presented. ECG analysis and interpretation program is developed. The delineation algorithm has been validated using QTDB. The results have been compared with those of other published approaches and have shown that the developed algorithm provides a reliable and accurate delineation of the ECG signal, outperforming other algorithms, and with errors well within the observed intercardiologist variations. The most significant improvement was found in the T wave and P wave delineation. ER -