Comparison Between Neural-Network-Based Adaptive Filtering and Wavelet Transform for ECG Characteristic Points Detection

As a physiologic response during exercise, due to tachycardia effects and liberated catecholamines, reduction of left ventricle volume is produced. Based on the Urotly effect, R Wave amplitude lessening during the stress tcst is reflected in the electrocardiographic signal. In spite of the previous statements, many researches during the last decades have been unable to find any ability of this parameter to assess Coronary Artery Disease (CAD) patients. The proposed methodology considering trend series analysis allows to approach several electrocardiographic parameters, hemodynamic responses and symptoms, allowing the correlation among them and to evaluate the behavior of each parameter. In a sample of 17 CAD patients and 14 healthy subjects assessed by either clustering analysis o r Mc Nemar test Iws shown adequate ability in group discrimination.