Session S43.2

A Method for Assessing the Sampling Bandwidth for Activation Time and Voltage Maps in Cardiac Navigators

JL Rojo-Alvarez*, FJ Sáiz-Rodríguez, CA Ruiz-Villa, F Alonso-Atienza, A García-Alberola

University Rey Juan Carlos
Fuenlabrada, Spain

Sequential Cardiac Navigation Systems (SCNS) are widely used to create an actual map of a heart chamber, by successively sampling the electrograms (EGM) inside the heart in a number of locations, whose spatial coordinates are also measured by a catheter location system. Post-processed EGM are subsequently used to create feature maps, which show either the spatial distribution of EGM activation times, yielding the depolarization impulse time course and decreased conduction velocity regions, or the maximum EGM voltage amplitudes, showing the regions with normal conduction and with scars. These maps are fundamental to determine the ablation targets. Given that the number or sampled EGM is a major determinant of the duration of the clinical ablation procedure, we propose a method suitable for analyzing the required spatial sampling rate in spatial maps. As cardiac chambers are asymmetrical, a simplification consisted on projecting each measured feature (activation time, amplitude of unipolar and bipolar EGM) into a 2-dimensional image, according to the azimuth and elevation coordinates of its spatial location (chamber gravity centre as coordinate origin). This non-uniformly sampled image was reinterpolated into a uniform image grid with k-nearest neighbour method adapted to spherical coordinates, and the bandwidth could be determined from a processed version of the 2D Fourier Transform. We used a conservative 10dB bandwidth, to keep the details for the clinical application. In order to test the suitability of the method, we analyzed 4 available cases of focal tachycardia (2) and flutter (2) in the right atria, from Carto (TM) SCNS. The number of EGM in the maps was in the rank 55-75. We analyzed the 10dB bandwidth of activation time and voltage maps, obtaining in average (2.19, 1.25, 1.48 rad^(-1)) for atrial focal tachycardia, and (2.34, 1.79, 2.19 rad^(-1)) for atrial flutter. Trends were observed for: (1) atrial flutter requiring higher bandwidths in all the features; (2) activation time maps being the most required bandwidth; (3) unipolar maps being the least required bandwidth. We conclude that the proposed method allows to analyze the spatial sampling requirements of current SCNS according to the arrhythmic substrate.

(Abstract Control Number: 75)