Session M1.2

Spatial Projection of Tachycardia Electrograms for Morphology Discrimination in Implantable Cardioverter Defibrillators

P Bouchet*, R Dubois, C Henry, P Roussel, G Dreyfus

Sorin Group CRM
Le Plessis-Robinson, France

Discrimination of Ventricular Tachycardia (VT) from Supra-Ventricular Tachycardia (SVT) remains a major challenge for appropriate therapy delivery in ICDs. Historically, only time intervals extracted from electrograms (EGMs) were used for diagnosis. In the last decade, morphology algorithms were added to improve performances, especially in single chamber devices. We propose a new discrimination algorithm that analyses the morphology of a two-dimensional representation of EGMs, named "Spatial Projection Of Tachycardia" (SPOT). The SPOT curve of a cardiac cycle is the plot of the amplitude of the far-field EGM versus the amplitude of the near-field EGM, with time as a parameter.
Morphological features are extracted from the comparison of arrhythmia and reference SPOT curves, based on physiological prior knowledge: the average angle of the relative velocity vectors, the correlation coefficient between the norms of the velocity and the correlation coefficient between their curvatures. A statistical classifier, specifically a Support Vector Machine, is trained with these three features and two additional timing features, computed on a training set of data. The classifier divides the feature space into two regions, giving the equation of the boundary surface as output. For arrhythmia classification, a simple computation determines the region to which the arrhythmia belongs, i.e. if whether it is a VT or an SVT.
Model training was performed from a private database: 29 induced VT and 19 induced SVT from 32 patients (57 ± 15.5 years, 87.5% men, 50% Ischemic Heart Disease). The procedure provided a classifier with 96.6% sensitivity and 94.7% specificity on that database. It was tested on the standard Ann Arbor Electrogram Libraries (AAEL): 64 VT and 7 SVT from 41 patients (61.9 ± 13.2 years, 82.9% men, 73.1% Coronary Artery Disease). On those fresh data, the classifier had 96.9% sensitivity and 85.7% specificity.
SPOT-based discrimination algorithm demonstrated high sensitivity and specificity for VT/SVT discrimination. The computational requirements of this algorithm make it appropriate for implementation in a single chamber ICD.

(Abstract Control Number: 84)