HRV signal dynamic extraction in the poincare plot by analyzing the extended U-Sequences for cardiac arrhythmia classification

This article presents a new method for extracting the dynamic of heart rate variability signal by studding the position of the points in the Poincare plot with respect to the line of identity for cardiac arrhythmia classification. In this method the points of HRV signal in Poincare plot are changed in to digital codes. Then, sequences of the codes are formed. At the end, by analyzing the manner of the repetitions and the continuance of these sequences in different arrhythmia classes, it could be possible to differentiate the arrhythmias from each other. One of the major advantages of this method is to discriminate arrhythmia by applying only 10 samples of RR intervals. To evaluate the performance of the proposed method, 3 different types of arrhythmia classes are utilized and they are classified with an accuracy of 95.6%, 93.04% and 97.21%, respectively.