Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learning

Abstract Functional connectivity (FC) analyses of intracranial EEG (iEEG) signals can potentially improve the mapping of epileptic networks in drug-resistant focal epilepsy. However, it remains unclear whether FC-based metrics provide additional value beyond established epilepsy biomarkers such as e...

Full description

Saved in:
Bibliographic Details
Main Authors: Jared Pilet, Scott A. Beardsley, Chad Carlson, Christopher T. Anderson, Candida Ustine, Sean Lew, Wade Mueller, Manoj Raghavan
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-02679-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849325995108073472
author Jared Pilet
Scott A. Beardsley
Chad Carlson
Christopher T. Anderson
Candida Ustine
Sean Lew
Wade Mueller
Manoj Raghavan
author_facet Jared Pilet
Scott A. Beardsley
Chad Carlson
Christopher T. Anderson
Candida Ustine
Sean Lew
Wade Mueller
Manoj Raghavan
author_sort Jared Pilet
collection DOAJ
description Abstract Functional connectivity (FC) analyses of intracranial EEG (iEEG) signals can potentially improve the mapping of epileptic networks in drug-resistant focal epilepsy. However, it remains unclear whether FC-based metrics provide additional value beyond established epilepsy biomarkers such as epileptic spikes and high-frequency oscillations (HFOs). Using interictal iEEG data from 26 patients, we estimated FC across eight frequency bands (4–290 Hz) using amplitude envelope correlation (AEC) and phase locking value (PLV). From the resulting FC-matrices, we estimated two graph metrics each to derive 32 FC-based features. We also extracted features related to spikes, HFOs, and power spectral densities (PSD). A trained support vector machine (SVM) classifier predicted seizure onset zones (SOZs) with an area under the ROC curve (AUC) of 0.91 for node-level 4-fold cross-validation (CV), 0.69 for patient-level 4-fold CV, and 0.73 for patient-level leave-one-out CV. Notably, gamma-band graph features from AECs outperformed spikes and HFOs in SOZ prediction when using an equivalent number of features. Our results strongly suggest that AEC-based features may provide more information about epileptogenicity compared to PLV-based features. Furthermore, machine learning provides a robust approach for identifying useful FC-based features and integrating information from putative biomarkers of epilepsy to better localize epileptogenic networks.
format Article
id doaj-art-c2caa8c6f3b945e2877c6ab08a19aa98
institution Kabale University
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-c2caa8c6f3b945e2877c6ab08a19aa982025-08-20T03:48:15ZengNature PortfolioScientific Reports2045-23222025-05-0115111510.1038/s41598-025-02679-4Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learningJared Pilet0Scott A. Beardsley1Chad Carlson2Christopher T. Anderson3Candida Ustine4Sean Lew5Wade Mueller6Manoj Raghavan7Joint Department of Biomedical Engineering, Marquette University and Medical College of WisconsinJoint Department of Biomedical Engineering, Marquette University and Medical College of WisconsinDepartment of Neurology, Medical College of WisconsinDepartment of Neurology, Medical College of WisconsinDepartment of Neurology, Medical College of WisconsinDepartment of Neurosurgery, Medical College of WisconsinDepartment of Neurosurgery, Medical College of WisconsinDepartment of Neurology, Medical College of WisconsinAbstract Functional connectivity (FC) analyses of intracranial EEG (iEEG) signals can potentially improve the mapping of epileptic networks in drug-resistant focal epilepsy. However, it remains unclear whether FC-based metrics provide additional value beyond established epilepsy biomarkers such as epileptic spikes and high-frequency oscillations (HFOs). Using interictal iEEG data from 26 patients, we estimated FC across eight frequency bands (4–290 Hz) using amplitude envelope correlation (AEC) and phase locking value (PLV). From the resulting FC-matrices, we estimated two graph metrics each to derive 32 FC-based features. We also extracted features related to spikes, HFOs, and power spectral densities (PSD). A trained support vector machine (SVM) classifier predicted seizure onset zones (SOZs) with an area under the ROC curve (AUC) of 0.91 for node-level 4-fold cross-validation (CV), 0.69 for patient-level 4-fold CV, and 0.73 for patient-level leave-one-out CV. Notably, gamma-band graph features from AECs outperformed spikes and HFOs in SOZ prediction when using an equivalent number of features. Our results strongly suggest that AEC-based features may provide more information about epileptogenicity compared to PLV-based features. Furthermore, machine learning provides a robust approach for identifying useful FC-based features and integrating information from putative biomarkers of epilepsy to better localize epileptogenic networks.https://doi.org/10.1038/s41598-025-02679-4Machine learningIntracranial EEGFunctional connectivitySeizure onset zoneEpileptogenicityBiomarkers
spellingShingle Jared Pilet
Scott A. Beardsley
Chad Carlson
Christopher T. Anderson
Candida Ustine
Sean Lew
Wade Mueller
Manoj Raghavan
Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learning
Scientific Reports
Machine learning
Intracranial EEG
Functional connectivity
Seizure onset zone
Epileptogenicity
Biomarkers
title Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learning
title_full Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learning
title_fullStr Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learning
title_full_unstemmed Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learning
title_short Predicting seizure onset zones from interictal intracranial EEG using functional connectivity and machine learning
title_sort predicting seizure onset zones from interictal intracranial eeg using functional connectivity and machine learning
topic Machine learning
Intracranial EEG
Functional connectivity
Seizure onset zone
Epileptogenicity
Biomarkers
url https://doi.org/10.1038/s41598-025-02679-4
work_keys_str_mv AT jaredpilet predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning
AT scottabeardsley predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning
AT chadcarlson predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning
AT christophertanderson predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning
AT candidaustine predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning
AT seanlew predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning
AT wademueller predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning
AT manojraghavan predictingseizureonsetzonesfrominterictalintracranialeegusingfunctionalconnectivityandmachinelearning