This dataset contains 100 models of the left atrium. Models come from 50 distinct patients, divided amongst 5 users to assess for inter- and intra-operator variability. The split used was 30 pairs (60 models) for inter-operator variability and 20 pairs (40 models) for intra operator variability. Models were created with a specific version of the software CemrgApp (cemrgapp.com), in which users processed a contrast enhanced magnetic resonance angiogram, and a late gadolinium enhanced (LGE) contrast magnetic resonance (CMR). Two types of simulations were run on each of the 100 processed cases: baseline pacing to calculate local activation time (LAT) maps and atrial fibrillation simulations for which phase singularity (PS) maps were calculated. The openCARP simulator (Plank et al., 2021) was used to run the simulations, using the Courtemance human atrial model with AF electrical remodelling. This dataset contains the labelled surface meshes, output to the CemrgApp software, which in turn are utilised as inputs for the electrophisiological simulations. The dataset is split into 100 folders labelled M1 to M100. Included in file `Cases_and_Users_Paths.csv` is the pairs for each of the comparisons, whether inter- or intra-observer variability.

{"references": ["Razeghi, et al. (2020). CemrgApp: An interactive medical imaging application with image processing, computer vision, and machine learning toolkits for cardiovascular research. DOI:10.1016/j.softx.2020.100570", "Plank, et al. (2021). The openCARP simulation environment for cardiac electrophysiology. DOI: 10.1016/j.cmpb.2021.106223."]}

Additional funding acknowledgements: British Heart Foundation (PG/15/91/31812, and PG/13/37/30280). Kings Health Partners London National Institute for Health Research (NIHR) Biomedical Research Centre. National Institute of Health (NIH) (R01-HL152256)