Spatial Distribution of Wilson's Central Terminal (WCT) on the Body Surface: Meshes, Body Surface Potential Maps, Atrial Activation Times, and WCT Analysis for 25 Torso Models and 100 Bi-Atrial Models

This dataset accompanies the publication: "Wilson's Central Terminal Changes Location on the Body Surface During the P-Wave: Why Precordial Leads Might Not Be What We Think" by Bender et al. (https://doi.org/10.64898/2026.05.20.26352966). Overview This dataset includes: body surface potential maps (BSPMs) 12-lead ECGs atrial activation times (LATs) models for a cohort of 100 atrial and 25 torso models derived from statistical shape models (SSMs) Atrial Models The bi-atrial SSM and the derived models are openly available under the Creative Commons license CC-BY 4.0 (https://doi.org/10.5281/zenodo.5571925) and presented in Nagel et al., “A Bi-Atrial Statistical Shape Model for Large-Scale in Silico Studies of Human Atria.” (https://doi.org/10.1016/j.media.2021.102210). Atrial models in this dataset are taken either from the original SSM cohort (100 random instances derived from the model, named cn617_gxxx.zip) or the LAE cohort (95 geometries with uniformly distributed left atrial volumes in LAE_geometries.zip). Note that the geometries from the LAE cohort were specifically selected here from those with volumes within the normal left atrial volume ranges. For details, see the description of the model dataset: https://doi.org/10.5281/zenodo.5571925. Torso Models The body SSM is openly available under the Simplified BSD License at http://humanshape.mpi-inf.mpg.de/. Torso models were derived by varying the two leading eigenvectors of the SSM systematically in a range of [−2𝜎, +2𝜎] as presented in Nagel et al., “Non-Invasive and Quantitative Estimation of Left Atrial Fibrosis Based on P Waves of the 12-Lead ECG—A Large-Scale Computational Study Covering Anatomical Variability.” (https://doi.org/10.3390/jcm10081797). Data Structure 1. Atrial Activation (/LATs) Stored as atrialModel.vtk files including: Material Tags: tags for atrial regions as in Nagel et al., “A Bi-Atrial Statistical Shape Model for Large-Scale in Silico Studies of Human Atria.” (https://doi.org/10.1016/j.media.2021.102210) / https://doi.org/10.5281/zenodo.5571925 Fibers: Preferential myocyte orientation vectors (‘fiber orientation’). LATs: Local Activation Times (measured in seconds). 2. Results (/results): All data is provided in MATLAB (.mat) format. Files follow a standardized naming convention: atrialModel_torsoModel.mat. Each .mat file contains: bspm: body surface potential maps (rows = points in torso model, columns = timesteps, time discretization 1ms; [nPointsTorso;nTimepoints]) ecg: 12-lead ECG with leads I,II,III, aVR, aVL, aVF, V1-V6: (rows = leads (12), columns = timesteps, time discretization 1ms; [12;nTimepoints]) percentageOfTime: percentage of time of the P-wave during which |BSPM-WCT| ≤ 5µV for each point on the torso surface ([1;nPointsTorso]) nPointsTorso is the number of points in the respective torso model and nTimesteps the number of timesteps. 3. Torso Geometry (/torsoModels) 25 torso models, T13 is the mean shape torsoModel.vtk: The 3D mesh of the torso. torsoModel_electrodes.csv: (x,y,z) coordinates for right arm, left arm, left foot, V1-V6, right foot ECG electrodes