The stroke lesion, 3D voxelwise probabilistic atlas was generated as follows: All ISLES’22 cases (n=400) were linearly registered to a FLAIR MNI template available in Schirmer et al. (2019) using NiftyReg (Ourselin et al. 2002). Afterwards, the binary segmentation masks were projected to the MNI space using the found (DWI to FLAIR) affine transformations. Each voxel from the 3D atlas counts the number of scans that showed infarcted tissue hitting the considered voxel (minimum number of scans 0, maximum 400). Format The atlas is shared using the compressed Nifty format. Citing the ISLES’22 stroke atlas For more information, please check our ISLES’22 preprint: https://arxiv.org/abs/2206.06694 If you use this atlas, please cite the following work: Hernandez Petzsche, M.R., de la Rosa, E., Hanning, U. et al. ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation dataset. Sci Data 9, 762 (2022). https://doi.org/10.1038/s41597-022-01875-5

{"references": ["Schirmer, Markus D., Anne-Katrin Giese, Panagiotis Fotiadis, Mark R. Etherton, Lisa Cloonan, Anand Viswanathan, Steven M. Greenberg, Ona Wu, and Natalia S. Rost. \"Spatial signature of white matter hyperintensities in stroke patients.\" Frontiers in neurology 10 (2019): 208.", "Ourselin, S\u00e9bastien, Radu Stefanescu, and Xavier Pennec. \"Robust registration of multi-modal images: towards real-time clinical applications.\" In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 140-147. Springer, Berlin, Heidelberg, 2002."]}