Catheter Ablation -> SVT/AVNRT/WPW/AT: -> Mapping & Imaging D-PO05 - Poster Session V (ID 39) Poster

D-PO05-167 - Validation Of An Algorithm For Automatic Arrhythmia Recognition And 3D Mapping In A Porcine Model (ID 576)

Abstract

Background: Atrial tachycardias (AT) can present multiple sites of origin or variable circuits which may complicate mapping and ablation, requiring the creation of separate 3D activation maps for each site.
Objective: To evaluate the ability of a novel Intra-Cardiac Pattern Matching (ICPM) software to automatically detect and assign different arrhythmia sources to separate 3D activation maps in a porcine model.
Methods: To simulate different AT sites, continuous pacing at same cycle length was done from 2-3 RA sites (2 screw-in leads and mapping catheter): 60-90 s of pacing and alternate to next site. RA was continuously mapped with an Octaray (48 electrodes, HD mapping catheter). Maps were created in standard fashion with an operator manually switching and adding points to the respective maps when the AT changed.
Automatic maps were created with the ICPM algorithm (Carto Mapping system). This analyzes different AT sites and automatically classifies each beat to its respective map. Unipolar EGM activations are matched against representative atrial activation patterns on the CS catheter in order to properly identify different arrhythmia sites.
In each animal, we then repositioned the pacing electrodes and created a second set of maps.
Offline analysis of manual and automatic maps (ICPM) was done using two methods: comparison of adjacent points (<5mm apart) from the 2 respective maps for their LAT values and Mesh coloring. Value differences of <10 msec were considered a match.
Results: In 6 pigs, a total of 22 different pacing sites were analyzed. For each site, 1 manual and 1 automatic map was created (44 maps). A total of 40176 points were compared between the manual automatic maps. Individual LAT for points on each map (manual and auto) were compared and found to match in 92.5% (variance of < 10ms). Mesh coloring values were also compared on the respective maps were also found to match using the same criteria.
Conclusion: The ICPM algorithm was able to accurately identify changing atrial activation sites and assign points to the appropriate map. A high correlation was found between the automatically identified and mapped atrial activation sites when compared to manual acquisitions.
Collapse