Fig 1.
Overview of 2D and 3D geometries of AF maps.
(A) Triangularized 2D grid with electrode splines highlighted in red. (B) Corresponding 3D triangularized mesh of the electrode basket, with electrode splines again highlighted in red. The triangle C4-D4-D5 shows the interpolated mesh after three subdivisions; this was omitted on other faces for clarity. (C) Corresponding atrial mesh determined from NavX or Carto, also with basket splines in red.
Table 1.
Patient demographics.
Fig 2.
Phase maps features are conserved between 2D and 3D.
(A) Snapshot of the 2D phase map in a 56 year old man with persistent AF showing a clockwise and counterclockwise rotational circuit. The phase singularities, corresponding to the tips of the spiral waves, are labeled in black (clockwise) and white (counterclockwise). (B) 3D electrode phase map corresponding to the timepoint in (A), showing the two rotational patterns seen in the 2D map. (C) Phase map projected onto the patient-specific anatomical atrial shell, along with the location of the spiral wave tips. (D-F) as in (A-C) showing three rotational circuits in a 71 year old man with persistent AF. (RSPV: Right superior pulmonary vein, RSPV: Right inferior pulmonary vein, LAA: Left atrial appendage, LPVs: Left pulmonary veins, LSPV: Left superior pulmonary vein).
Fig 3.
Rotational activation is conserved between 2D and reconstructed 3D shells.
(A) Snapshot of a phase map of the left atrium for a 30 year old man using the electrode grid mesh showing two phase singularities (white circles). (B) Phase singularities calculated using the 2D phase map (pink circles) and using the 3D phase map, projected onto the 2D grid. (C) As in (A), and (D) as in (B), showing the right atrium for a 47 year old woman.
Fig 4.
Tip density maps are conserved between 2D and 3D.
The tip or phase singularity density map computed using 2D phase maps (left) and 3D maps, projected onto 2D (right) for the patient of Fig 2A–2C. Both maps are virtually identical with a correlation coefficient of r = 0.994. Percent values represent the proportion of frames during which a given electrode grid contained a tip.
Fig 5.
Correlation coefficient of 2D and 3D tip density does not depend on tip location or shell curvature.
Plotted are, for all patients, the correlation coefficient between the 2D and 3D tip density maps for all electrodes (Full), electrodes located at the poles of the basket (Pole), the equator (Equator), and for the 7 electrodes with maximal (Max Curv) and minimal (Min Curv) mean curvature. The red bar represents the median value for each data set.