Detection of focal cortical dysplasia: Development and multicentric evaluation of artificial intelligence models
This study shows the variance in performance for FCD detection models in a multicenter dataset. The two models with 3D input data showed the highest sensitivity. The 2D models performed worse than all other models, suggesting that FCD detection requires 3D data. The greatly improved precision of 3D-nnUNet may make it a sensible choice to aid FCD detection.
Authors
- Bastian David (ORCID: https://orcid.org/0000-0002-0146-0629)
- Hans‐Jürgen Huppertz (ORCID: https://orcid.org/0000-0003-3856-9094)
- Vadym Gnatkovsky (ORCID: https://orcid.org/0000-0002-4543-0464)
- Felix Rosenow (ORCID: https://orcid.org/0000-0002-3989-7471)
- Fabiane Schuch (ORCID: https://orcid.org/0000-0002-5832-9904)
- Tobias Bauer (ORCID: https://orcid.org/0000-0002-0555-6214)
- Lennart Walger (ORCID: https://orcid.org/0000-0002-3300-6877)
- Fee Christiane Keil (ORCID: https://orcid.org/0000-0002-9158-1538)
- Angela M. Kaindl (ORCID: https://orcid.org/0000-0001-9454-206X)
- Elke Hattingen (ORCID: https://orcid.org/0000-0002-8392-9004)
- Rainer Surges (ORCID: https://orcid.org/0000-0002-3177-8582)
- Alexander Radbruch (ORCID: https://orcid.org/0000-0001-6238-6525)
- Anna Tietze (ORCID: https://orcid.org/0000-0002-2601-9055)
- Elisabeth Neuhaus (ORCID: https://orcid.org/0000-0002-7506-0169)
- Theodor Rüber (ORCID: https://orcid.org/0000-0002-6180-7671)
- Lennart Nils Kersting (ORCID: https://orcid.org/0009-0002-1983-4892)
Institutions
- Goethe University Frankfurt (DE)
- University of Bonn (DE)
- University Hospital Bonn (DE)
- German Center for Neurodegenerative Diseases (DE)
- Swiss Epilepsy Center (CH)
- Charité - Universitätsmedizin Berlin (DE)
Publication Details
- Journal
- Epilepsia
- Published
- 2024-12-31
- DOI
- https://doi.org/10.1111/epi.18240
- Citations
- 9
- Primary Topic
- Brain Tumor Detection and Classification
- Type
- article
- Field-Weighted Citation Impact
- 2.23