Geometric patterns of time-delay plots from different cardiac rhythms and arrhythmias using short-term EKG signals

To date, no systematic work has been intended to describe spatio-temporal patterns of cardiac rhythms using only short series of RR intervals, to facilitate visual or computerized-aided identification of EKG motifs for use in clinical practice. The aim of this study was to detect and classify eye-catching geometric patterns of Poincaré time-delay plots from different types of cardiac rhythms and arrhythmias using short-term EKG signals.

Approximately 150-300 representative, consecutive beats were retrieved from 24-h Holter registers of 100 patients with different heart rhythms. Two-dimensional Poincaré charts were created, and the resulting geometric patterns were transformed into representative familiar eye-catching drawings to interpret different arrhythmias.

Poincaré plot representation of RR interval data revealed a wide variety of visual patterns: (i) comet-shaped for sinus rhythm; (ii) torpedo-shaped for sinus bradycardia; (iii) cigarette-shaped for sinus tachycardia; (iv) butterfly-shaped for sinus tachycardia and isolated atrial premature complexes; (v) arrow-shaped for isolated premature complexes and inappropriate sinus tachycardia; (vi) inverted fan-shaped for sinus rhythm with frequent atrial premature complexes; (vii) tornado-shaped for atrial flutter and atrial tachycardia; and (viii) fan-shaped for atrial fibrillation.

Modified Poincaré plots with smoothed lines connecting successive points could accurately classify different types of arrhythmias based on short RR interval sequence variability. Characteristic emergent patterns can be visually identified and eventually could be distinguished by an automatic classification system able to discern between arrhythmias. This work provides an alternative method to interpret time-delay plots obtained from short-term EKG signal recordings.