000278020 001__ 278020
000278020 005__ 20250430100242.0
000278020 0247_ $$2doi$$a10.1111/epi.18240
000278020 0247_ $$2pmid$$apmid:39739580
000278020 0247_ $$2pmc$$apmc:PMC11997906
000278020 0247_ $$2ISSN$$a0013-9580
000278020 0247_ $$2ISSN$$a1528-1167
000278020 037__ $$aDZNE-2025-00530
000278020 041__ $$aEnglish
000278020 082__ $$a610
000278020 1001_ $$00009-0002-1983-4892$$aKersting, Lennart N$$b0
000278020 245__ $$aDetection of focal cortical dysplasia: Development and multicentric evaluation of artificial intelligence models.
000278020 260__ $$aOxford [u.a.]$$bWiley-Blackwell$$c2025
000278020 3367_ $$2DRIVER$$aarticle
000278020 3367_ $$2DataCite$$aOutput Types/Journal article
000278020 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1744812069_30974
000278020 3367_ $$2BibTeX$$aARTICLE
000278020 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000278020 3367_ $$00$$2EndNote$$aJournal Article
000278020 520__ $$aFocal 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% male) and 85 healthy controls (7.1 ± 2.4 years old, 51% female). We used 118 subjects for model training and kept the remaining subjects as an independent test set. 3D-nnUNet achieved the highest F1 score of .58, the highest DSC of .36 (95% confidence interval [CI] = .30-.41), a detection rate of 55%, and a specificity of 86%. deepFCD showed the highest detection rate (82%) but had the lowest specificity (0%) and cluster-level precision (.03, 95% CI = .03-.04, F1 score = .07). MELD showed the least performance variation across centers, with detection rates between 46% and 54%.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.
000278020 536__ $$0G:(DE-HGF)POF4-353$$a353 - Clinical and Health Care Research (POF4-353)$$cPOF4-353$$fPOF IV$$x0
000278020 536__ $$0G:(DE-HGF)POF4-354$$a354 - Disease Prevention and Healthy Aging (POF4-354)$$cPOF4-354$$fPOF IV$$x1
000278020 588__ $$aDataset connected to CrossRef, PubMed, , Journals: pub.dzne.de
000278020 650_7 $$2Other$$aMRI
000278020 650_7 $$2Other$$acomputer‐aided detection
000278020 650_7 $$2Other$$aepilepsy
000278020 650_7 $$2Other$$alesion detection
000278020 650_7 $$2Other$$amodel comparison
000278020 650_2 $$2MeSH$$aHumans
000278020 650_2 $$2MeSH$$aMale
000278020 650_2 $$2MeSH$$aFemale
000278020 650_2 $$2MeSH$$aMalformations of Cortical Development: diagnostic imaging
000278020 650_2 $$2MeSH$$aAdult
000278020 650_2 $$2MeSH$$aMagnetic Resonance Imaging: methods
000278020 650_2 $$2MeSH$$aRetrospective Studies
000278020 650_2 $$2MeSH$$aYoung Adult
000278020 650_2 $$2MeSH$$aChild
000278020 650_2 $$2MeSH$$aArtificial Intelligence
000278020 650_2 $$2MeSH$$aAdolescent
000278020 650_2 $$2MeSH$$aImaging, Three-Dimensional
000278020 650_2 $$2MeSH$$aFocal Cortical Dysplasia
000278020 7001_ $$00000-0002-3300-6877$$aWalger, Lennart$$b1
000278020 7001_ $$0P:(DE-2719)9002598$$aBauer, Tobias$$b2
000278020 7001_ $$aGnatkovsky, Vadym$$b3
000278020 7001_ $$aSchuch, Fabiane$$b4
000278020 7001_ $$0P:(DE-2719)9001570$$aDavid, Bastian$$b5
000278020 7001_ $$aNeuhaus, Elisabeth$$b6
000278020 7001_ $$aKeil, Fee$$b7
000278020 7001_ $$aTietze, Anna$$b8
000278020 7001_ $$aRosenow, Felix$$b9
000278020 7001_ $$00000-0001-9454-206X$$aKaindl, Angela M$$b10
000278020 7001_ $$aHattingen, Elke$$b11
000278020 7001_ $$aHuppertz, Hans-Jürgen$$b12
000278020 7001_ $$0P:(DE-2719)9001861$$aRadbruch, Alexander$$b13$$udzne
000278020 7001_ $$aSurges, Rainer$$b14
000278020 7001_ $$00000-0002-6180-7671$$aRüber, Theodor$$b15$$eLast author
000278020 773__ $$0PERI:(DE-600)2002194-X$$a10.1111/epi.18240$$gVol. 66, no. 4, p. epi.18240$$n4$$p1165 - 1176$$tEpilepsia$$v66$$x0013-9580$$y2025
000278020 8564_ $$uhttps://pub.dzne.de/record/278020/files/DZNE-2025-00530.pdf$$yOpenAccess
000278020 8564_ $$uhttps://pub.dzne.de/record/278020/files/DZNE-2025-00530.pdf?subformat=pdfa$$xpdfa$$yOpenAccess
000278020 909CO $$ooai:pub.dzne.de:278020$$popenaire$$popen_access$$pVDB$$pdriver$$pdnbdelivery
000278020 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)9002598$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b2$$kDZNE
000278020 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)9001861$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b13$$kDZNE
000278020 9101_ $$0I:(DE-588)1065079516$$60000-0002-6180-7671$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b15$$kDZNE
000278020 9131_ $$0G:(DE-HGF)POF4-353$$1G:(DE-HGF)POF4-350$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lNeurodegenerative Diseases$$vClinical and Health Care Research$$x0
000278020 9131_ $$0G:(DE-HGF)POF4-354$$1G:(DE-HGF)POF4-350$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lNeurodegenerative Diseases$$vDisease Prevention and Healthy Aging$$x1
000278020 9141_ $$y2025
000278020 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)1050$$2StatID$$aDBCoverage$$bBIOSIS Previews$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)1190$$2StatID$$aDBCoverage$$bBiological Abstracts$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bEPILEPSIA : 2022$$d2024-12-09
000278020 915__ $$0LIC:(DE-HGF)CCBYNC4$$2HGFVOC$$aCreative Commons Attribution-NonCommercial CC BY-NC 4.0
000278020 915__ $$0StatID:(DE-HGF)3001$$2StatID$$aDEAL Wiley$$d2024-12-09$$wger
000278020 915__ $$0StatID:(DE-HGF)1030$$2StatID$$aDBCoverage$$bCurrent Contents - Life Sciences$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000278020 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bEPILEPSIA : 2022$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)1110$$2StatID$$aDBCoverage$$bCurrent Contents - Clinical Medicine$$d2024-12-09
000278020 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2024-12-09
000278020 9201_ $$0I:(DE-2719)5000075$$kAG Radbruch$$lClinical Neuroimaging$$x0
000278020 9201_ $$0I:(DE-2719)1013026$$kAG Stöcker$$lMR Physics$$x1
000278020 980__ $$ajournal
000278020 980__ $$aVDB
000278020 980__ $$aUNRESTRICTED
000278020 980__ $$aI:(DE-2719)5000075
000278020 980__ $$aI:(DE-2719)1013026
000278020 9801_ $$aFullTexts