Prediction of Ventricular Tachycardia using Nonlinear Features of Heart Rate Variability Signal such as Poincare Plot, Approximate and Sample Entropy, Recurrence Plot

In the ventricular tachycardia (VT), due to improper contractions of the ventricles and excessive increase in heart rate, very little blood is released from the heart, and if not treated promptly, it can lead to the death of the patient. The occurrence of VT and its timely diagnosis are signs of heart rate changes (HRV) that are helpful in detecting it. But before it happens, it's not easy to find such symptoms, and this becomes even more acute when it comes to predicting the time to go backwards. The utilized data are taken from Physionet database The data studied in this article is data available in the MVTDB database of physionet, which includes 212 records from the patient group with VT and the control group. In this study, an algorithm was proposed to predict VT based on the extraction of nonlinear characteristics of the HRV signal. To evaluate the effectiveness of the features, t-test analysis was used and PCA algorithm was used to reduce the dimensions of the feature. The features that have been given to predict separation between two healthy and patient classes are given to the Artificial Neural Network(ANN). In this study, three different modes were studied to examine the values of the characteristics and how changes in their values at various time intervals could be a warning to the attack, and the results of all three modes were compared together and finally, with significant changes The values of the properties in the range of 130 to 10 seconds before the start of the attack, VT prediction with accuracy of 94.28% in this interval.