Detail
Článek
Článek online
FT
Medvik - BMČ
  • Je něco špatně v tomto záznamu ?

Leveraging interictal multimodal features and graph neural networks for automated planning of epilepsy surgery

P. Nejedly, V. Hrtonova, M. Pail, J. Cimbalnik, P. Daniel, V. Travnicek, I. Dolezalova, F. Mivalt, V. Kremen, P. Jurak, GA. Worrell, B. Frauscher, P. Klimes, M. Brazdil

. 2025 ; 7 (3) : fcaf140. [pub] 20250416

Status neindexováno Jazyk angličtina Země Anglie, Velká Británie

Typ dokumentu časopisecké články

Perzistentní odkaz   https://www.medvik.cz/link/bmc25014533

Precise localization of the epileptogenic zone is pivotal for planning minimally invasive surgeries in drug-resistant epilepsy. Here, we present a graph neural network (GNN) framework that integrates interictal intracranial EEG features, electrode topology, and MRI features to automate epilepsy surgery planning. We retrospectively evaluated the model using leave-one-patient-out cross-validation on a dataset of 80 drug-resistant epilepsy patients treated at St. Anne's University Hospital (Brno, Czech Republic), comprising 31 patients with good postsurgical outcomes (Engel I) and 49 with poor outcomes (Engel II-IV). The GNN predictions demonstrated a significantly better (P < 0.05, Mann-Whitney-U test) area under the precision-recall curve in patients with good outcomes (area under the precision-recall curve: 0.69) compared with those with poor outcomes (area under the precision-recall curve: 0.33), indicating that the model captures clinically relevant targets in successful cases. In patients with poor outcomes, the graph neural network proposed alternative intervention sites that diverged from the original clinical plans, highlighting its potential to identify alternative therapeutic targets. We show that topology-aware GNNs significantly outperformed (P < 0.05, Wilcoxon signed-rank test) traditional neural networks while using the same intracranial EEG features, emphasizing the importance of incorporating implantation topology into predictive models. These findings uncover the potential of GNNs to automatically suggest targets for epilepsy surgery, which can assist the clinical team during the planning process.

Citace poskytuje Crossref.org

000      
00000naa a2200000 a 4500
001      
bmc25014533
003      
CZ-PrNML
005      
20250905141421.0
007      
ta
008      
250701s2025 enk f 000 0|eng||
009      
AR
024    7_
$a 10.1093/braincomms/fcaf140 $2 doi
035    __
$a (PubMed)40357016
040    __
$a ABA008 $b cze $d ABA008 $e AACR2
041    0_
$a eng
044    __
$a enk
100    1_
$a Nejedly, Petr $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic $u Institute of Scientific Instruments, The Czech Academy of Sciences, Brno 612 00, Czech Republic
245    10
$a Leveraging interictal multimodal features and graph neural networks for automated planning of epilepsy surgery / $c P. Nejedly, V. Hrtonova, M. Pail, J. Cimbalnik, P. Daniel, V. Travnicek, I. Dolezalova, F. Mivalt, V. Kremen, P. Jurak, GA. Worrell, B. Frauscher, P. Klimes, M. Brazdil
520    9_
$a Precise localization of the epileptogenic zone is pivotal for planning minimally invasive surgeries in drug-resistant epilepsy. Here, we present a graph neural network (GNN) framework that integrates interictal intracranial EEG features, electrode topology, and MRI features to automate epilepsy surgery planning. We retrospectively evaluated the model using leave-one-patient-out cross-validation on a dataset of 80 drug-resistant epilepsy patients treated at St. Anne's University Hospital (Brno, Czech Republic), comprising 31 patients with good postsurgical outcomes (Engel I) and 49 with poor outcomes (Engel II-IV). The GNN predictions demonstrated a significantly better (P < 0.05, Mann-Whitney-U test) area under the precision-recall curve in patients with good outcomes (area under the precision-recall curve: 0.69) compared with those with poor outcomes (area under the precision-recall curve: 0.33), indicating that the model captures clinically relevant targets in successful cases. In patients with poor outcomes, the graph neural network proposed alternative intervention sites that diverged from the original clinical plans, highlighting its potential to identify alternative therapeutic targets. We show that topology-aware GNNs significantly outperformed (P < 0.05, Wilcoxon signed-rank test) traditional neural networks while using the same intracranial EEG features, emphasizing the importance of incorporating implantation topology into predictive models. These findings uncover the potential of GNNs to automatically suggest targets for epilepsy surgery, which can assist the clinical team during the planning process.
590    __
$a NEINDEXOVÁNO
655    _2
$a časopisecké články $7 D016428
700    1_
$a Hrtonova, Valentina $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic $u Institute of Scientific Instruments, The Czech Academy of Sciences, Brno 612 00, Czech Republic $u Department of Neurology, Duke University Medical Center, Durham, NC 27705, USA
700    1_
$a Pail, Martin $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic $u Institute of Scientific Instruments, The Czech Academy of Sciences, Brno 612 00, Czech Republic
700    1_
$a Cimbalnik, Jan $u International Clinical Research Center, St. Anne's University Hospital, Brno 602 00, Czech Republic
700    1_
$a Daniel, Pavel $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic
700    1_
$a Travnicek, Vojtech $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic $u Institute of Scientific Instruments, The Czech Academy of Sciences, Brno 612 00, Czech Republic $u Department of Neurology, Duke University Medical Center, Durham, NC 27705, USA
700    1_
$a Dolezalova, Irena $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic
700    1_
$a Mivalt, Filip $u Department of Neurology, Mayo Clinic, Rochester, MN 55905, USA $1 https://orcid.org/0000000206939495
700    1_
$a Kremen, Vaclav $u Department of Neurology, Mayo Clinic, Rochester, MN 55905, USA $u Department of Physiology and Biomedical Engineering, Mayo Clinic, Rochester, MN 55905, USA $1 https://orcid.org/0000000198447617
700    1_
$a Jurak, Pavel $u Institute of Scientific Instruments, The Czech Academy of Sciences, Brno 612 00, Czech Republic
700    1_
$a Worrell, Gregory A $u Department of Neurology, Mayo Clinic, Rochester, MN 55905, USA $u Department of Physiology and Biomedical Engineering, Mayo Clinic, Rochester, MN 55905, USA
700    1_
$a Frauscher, Birgit $u Department of Neurology, Duke University Medical Center, Durham, NC 27705, USA $u Department of Biomedical Engineering, Pratt School of Engineering, Duke University, Durham, NC 27705, USA
700    1_
$a Klimes, Petr $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic $u Institute of Scientific Instruments, The Czech Academy of Sciences, Brno 612 00, Czech Republic
700    1_
$a Brazdil, Milan $u Brno Epilepsy Center, Department of Neurology, Member of ERN-EpiCARE, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno 625 00, Czech Republic
773    0_
$w MED00205536 $t Brain communications $x 2632-1297 $g Roč. 7, č. 3 (2025), s. fcaf140
856    41
$u https://pubmed.ncbi.nlm.nih.gov/40357016 $y Pubmed
910    __
$a ABA008 $b sig $c sign $y - $z 0
990    __
$a 20250701 $b ABA008
991    __
$a 20250905141408 $b ABA008
999    __
$a ok $b bmc $g 2388114 $s 1251653
BAS    __
$a 3
BAS    __
$a PreBMC-PubMed-not-MEDLINE
BMC    __
$a 2025 $b 7 $c 3 $d fcaf140 $e 20250416 $i 2632-1297 $m Brain communications $n Brain Commun $x MED00205536
LZP    __
$a Pubmed-20250701
Zpět

Najít záznam

Citační ukazatele

Možnosti archivace