Detection of cardiac arryhthmia by using time domaine technique on heart rate variabilty signals

The analysis of the Variability of the Heart Rate (HRV) is coming as an important indicator for different clinical applications like the prediction of arrhythmias, sudden cardiac death, assessing cardiovascular and metabolic illness progression or in sports physiology. In this paper we have developed an algorithm to detect a supraventricular arrhythmia, by processing the heart rate variability (HRV) signals. For this, we used time domain method for features extraction and MLP and RBF neural network for classification. To evaluate our system proposed two international standard databases namely MIT-BIH Supraventricular Arrhythmia Database and Normal Sinus Rhythm RR Intervals Database were used. We estimate that the results obtained are very satisfactory since the MLP and RBF accuracy located between 85% and 100%.