Detection of ventricular fibrillation using the autocorrelation function analysis of the ECG

A method is developed for the detection of ventricular fibrillation (VF) and life-threatening arrhythmias. The method is based on direct and simple peak analysis of the autocorrelation function (ACF). It can differentiate between VF (coarse and fine) and non-VH rhythms. ECG records during ventricular tachycardia (VT) and VF were obtained, and 4-s-long segments were digitized at 200 Hz and then split in three groups (VT, VF regular, and VF irregular). ACFs were computed, and positive peak P(j) (j=1, 2, . . .), RPL(j)=P(j)/2SE(1), TR(1)=P(1) width/lag P(1), and D(j)=P(j)-P(j+1) were measured and calculated for each sample. Results show that: (a) RPL(j)(j=1, 2, 3) together with D(j) present high specificity to discriminate between VT and VF; and (b) TR(1) and those parameters make it possible to discriminate between SVT and ventricular-originated tachyarrhythmias. The method is reliable and simple.<
>