Font Size: 

Cardiac arrhythmia is a common clinical problem in cardiology definedas the abnormality in heart rhythm. Bradycardia, atrial fibrillation, tachycardia,supraventricular tachycardia, atrial flutter and sinus irregularity are commondifferent classifications for arrhythmia. Here, we present a new approach todistinguishing between these most common heart rhythms. Our approach is basedon dynamical system techniques, namely recurrence entropy of microstates, andrecurrence vicinity threshold, in association with artificial intelligence. The rhythmsand other cardiac conditions of the dataset were labeled by more than one licensedphysician. The main contributions of this work are the identification of how differentheart rhythms affects the entropy of recurrence microstates and recurrence vicinitythreshold parameter, and in doing so, this quantifier may be used as a featureextraction to artificial intelligence classifiers. Our method involves a significantreduction of the data set to be analyzed by machine learning algorithms and canbring benefits in situations of pre-testing individuals, due to the minimumprocessing time and hardware required to perform the analysis. The additionalinformation obtained by the two quantifiers may also be put together with the signals, consolidating data from multiple sources, adding more useful informationto the dataset.