%0 Journal Article
%A Kersting, Lennart N
%A Walger, Lennart
%A Bauer, Tobias
%A Gnatkovsky, Vadym
%A Schuch, Fabiane
%A David, Bastian
%A Neuhaus, Elisabeth
%A Keil, Fee
%A Tietze, Anna
%A Rosenow, Felix
%A Kaindl, Angela M
%A Hattingen, Elke
%A Huppertz, Hans-Jürgen
%A Radbruch, Alexander
%A Surges, Rainer
%A Rüber, Theodor
%T Detection of focal cortical dysplasia: Development and multicentric evaluation of artificial intelligence models.
%J Epilepsia
%V 66
%N 4
%@ 0013-9580
%C Oxford [u.a.]
%I Wiley-Blackwell
%M DZNE-2025-00530
%P 1165 - 1176
%D 2025
%X Focal cortical dysplasia (FCD) is a common cause of drug-resistant focal epilepsy but can be challenging to detect visually on magnetic resonance imaging. Three artificial intelligence models for automated FCD detection are publicly available (MAP18, deepFCD, MELD) but have only been compared on single-center data. Our first objective is to compare them on independent multicenter test data. Additionally, we train and compare three new models and make them publicly available.We retrospectively collected FCD cases from four epilepsy centers. We chose three novel models that take two-dimensional (2D) slices (2D-nnUNet), 2.5D slices (FastSurferCNN), and large 3D patches (3D-nnUNet) as inputs and trained them on a subset of Bonn data. As core evaluation metrics, we used voxel-level Dice similarity coefficient (DSC), cluster-level F1 score, subject-level detection rate, and specificity.We collected 329 subjects, 244 diagnosed with FCD (27.7 ± 14.4 years old, 54
%K Humans
%K Male
%K Female
%K Malformations of Cortical Development: diagnostic imaging
%K Adult
%K Magnetic Resonance Imaging: methods
%K Retrospective Studies
%K Young Adult
%K Child
%K Artificial Intelligence
%K Adolescent
%K Imaging, Three-Dimensional
%K Focal Cortical Dysplasia
%K MRI (Other)
%K computer‐aided detection (Other)
%K epilepsy (Other)
%K lesion detection (Other)
%K model comparison (Other)
%F PUB:(DE-HGF)16
%9 Journal Article
%$ pmid:39739580
%2 pmc:PMC11997906
%R 10.1111/epi.18240
%U https://pub.dzne.de/record/278020