Effects of compensatory aggregation operators on fuzzy combination of evidence in a QRS detector

A system for the detection of QRS in ECG is developed that uses fuzzy logic to combine the outputs of two QRS detector algorithms. Fuzzy combination of the evidence provided by two algorithms with different sensitivities to noise is used to determine the time points at which QRS complexes occur. The fuzzy detector incorporates a measure of uncertainty and can conclude that a QRS is present that a QRS is not present or that no decision can be made. The fuzzy system was implemented and evaluated using several methods of fuzzy aggregation, including the minimum operator, noncompensatory operators, and compensatory operators. For the minimum operator, error rates were substantially reduced below those obtained with the individual algorithms, but at a cost of a large number of uncertain detections. A highly compensated operator produced very few uncertain detections and fewer errors than the individual algorithms, but the error rates were larger than those obtained with the minimum operator. The advantage to identifying uncertain time points is that a small number of points is selected for additional analysis and these points are better a priori evidence, compared to an arbitrary time point in the waveform, that a QRS complex exists