Abstract Introduction There exist many imaging techniques and systems to reproduce atrial chambers in 3D. These technologies include electroanatomical (EA) mapping systems, noninvasive electrocardiographic imaging (ECGI), magnetic resonance imaging (MRI), or computed tomography (CT) scans. In the case of atrial fibrillation (AF), the most employed non-pharmacological treatment is catheter ablation to electrically isolate the pulmonary veins from the rest of the left atrium. Driver mechanisms such as focal or rotational activity have been proposed as possible initiating and maintaining mechanisms of AF. However, correspondence and validation of these sites when several systems are employed in the same patient remains a challenge, as they are mostly manually aligned based on visual inspection. Purpose To develop an automatic 3D alignment algorithm for cardiac 3D meshes to colocalize points between atrial maps generated with multiple EA mapping systems, ECGI, MRI, or CT scans. Methods A total of 25 left atrial meshes from persistent AF patients were exported from an EA mapping system. The total number of vertices for all the meshes was 2545444 points (101817.8±13593.3 points per map). A reference mesh was employed with minor modifications [1]. All meshes were manually segmented into 12 different left atrial regions, see Table for the region names. The method implements a non-rigid variant of the iterative closest point algorithm to transform the atrial mesh onto the reference one, see Figure. The geographical distance between the mean position of the 12 different segmented reference areas and the 12 transformed points was employed as the performance metric. Results The global error for all the fiducial points in all left atrial meshes was 11.57±2.55 mm. The average local errors for the 12 atrial areas are summarized in the Table. The best three aligned areas were the RSPV, atrial septum, and lateral wall. The areas with less alignment accuracy were the LAA, LSPV, and atrial roof. Conclusions The algorithm provides a promising solution to evaluate and validate site-related results from different systems, e.g., rotational activity presence between EA mapping and ECGI systems. The method works automatically for any given chamber anatomy or any number of points. No prior segmentation is needed since the transformation and co-localization are applied to the raw chamber mesh. Further analysis with a larger mesh database is needed. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Instituto de Salud Carlos III and Ministerio de Ciencia, Innovaciόn y Universidades