Abstract

Objective: Epilepsy surgery results in seizure freedom in the majority of drug-resistant patients. To improve surgery outcome we studied whether MEG metrics combined with machine learning can improve localization of the epileptogenic zone, thereby enhancing the chance of seizure freedom. Methods: Presurgical interictal MEG recordings of 94 patients (64 seizure-free > 1y post-surgery) were analyzed to extract four metrics in source space: delta power, low-to-high-frequency power ratio, functional connectivity (phase lag index), and minimum spanning tree betweenness centrality. At the group level, we estimated the overlap of the resection area with the five highest values for each metric and determined whether this overlap differed between surgery outcomes. At the individual level, those metrics were used in machine learning classifiers (linear support vector machine (SVM) and random forest) to distinguish between resection and non-resection areas and between surgery outcome groups. Results: The highest values, for all metrics, overlapped with the resection area in more than half of the patients, but the overlap did not differ between surgery outcome groups. The classifiers distinguished the resection areas from non-resection areas with 59.94% accuracy (95% confidence interval: 59.67-60.22%) for SVM and 60.34% (59.98-60.71%) for random forest, but could not differentiate seizure-free from not seizure-free patients [43.77% accuracy (42.08-45.45%) for SVM and 49.03% (47.25-50.82%) for random forest]. Significance: All four metrics localized the resection area but did not distinguish between surgery outcome groups, demonstrating that metrics derived from interictal MEG correspond to expert consensus based on several presurgical evaluation modalities, but do not yet localize the epileptogenic zone. Metrics should be improved such that they correspond to the resection area in seizure-free patients but not in patients with persistent seizures. It is important to test such localization strategies at an individual level, for example by using machine learning or individualized models, since surgery is individually tailored.
Original languageEnglish
Article number647
JournalFrontiers in Neurology
Volume9
Issue numberAUG
DOIs
Publication statusPublished - 7 Aug 2018

Cite this

@article{40b570fd703044999f998c91553cb40c,
title = "Localization of the epileptogenic zone using interictal MEG and machine learning in a large cohort of drug-resistant epilepsy patients",
abstract = "Objective: Epilepsy surgery results in seizure freedom in the majority of drug-resistant patients. To improve surgery outcome we studied whether MEG metrics combined with machine learning can improve localization of the epileptogenic zone, thereby enhancing the chance of seizure freedom. Methods: Presurgical interictal MEG recordings of 94 patients (64 seizure-free > 1y post-surgery) were analyzed to extract four metrics in source space: delta power, low-to-high-frequency power ratio, functional connectivity (phase lag index), and minimum spanning tree betweenness centrality. At the group level, we estimated the overlap of the resection area with the five highest values for each metric and determined whether this overlap differed between surgery outcomes. At the individual level, those metrics were used in machine learning classifiers (linear support vector machine (SVM) and random forest) to distinguish between resection and non-resection areas and between surgery outcome groups. Results: The highest values, for all metrics, overlapped with the resection area in more than half of the patients, but the overlap did not differ between surgery outcome groups. The classifiers distinguished the resection areas from non-resection areas with 59.94{\%} accuracy (95{\%} confidence interval: 59.67-60.22{\%}) for SVM and 60.34{\%} (59.98-60.71{\%}) for random forest, but could not differentiate seizure-free from not seizure-free patients [43.77{\%} accuracy (42.08-45.45{\%}) for SVM and 49.03{\%} (47.25-50.82{\%}) for random forest]. Significance: All four metrics localized the resection area but did not distinguish between surgery outcome groups, demonstrating that metrics derived from interictal MEG correspond to expert consensus based on several presurgical evaluation modalities, but do not yet localize the epileptogenic zone. Metrics should be improved such that they correspond to the resection area in seizure-free patients but not in patients with persistent seizures. It is important to test such localization strategies at an individual level, for example by using machine learning or individualized models, since surgery is individually tailored.",
keywords = "magnetoencephalography, presurgical evaluation, functional connectivity, refractory epilepsy, seizure freedom, beamforming",
author = "Nissen, {Ida A.} and Stam, {Cornelis J.} and {van Straaten}, {Elisabeth C. W.} and Viktor Wottschel and Reijneveld, {Jaap C.} and Baayen, {Johannes C.} and Hamer, {Philip C. de Witt} and Sander Idema and Velis, {Demetrios N.} and Arjan Hillebrand",
year = "2018",
month = "8",
day = "7",
doi = "10.3389/fneur.2018.00647",
language = "English",
volume = "9",
journal = "Frontiers in Neurology",
issn = "1664-2295",
publisher = "Frontiers Media S.A.",
number = "AUG",

}

TY - JOUR

T1 - Localization of the epileptogenic zone using interictal MEG and machine learning in a large cohort of drug-resistant epilepsy patients

AU - Nissen, Ida A.

AU - Stam, Cornelis J.

AU - van Straaten, Elisabeth C. W.

AU - Wottschel, Viktor

AU - Reijneveld, Jaap C.

AU - Baayen, Johannes C.

AU - Hamer, Philip C. de Witt

AU - Idema, Sander

AU - Velis, Demetrios N.

AU - Hillebrand, Arjan

PY - 2018/8/7

Y1 - 2018/8/7

N2 - Objective: Epilepsy surgery results in seizure freedom in the majority of drug-resistant patients. To improve surgery outcome we studied whether MEG metrics combined with machine learning can improve localization of the epileptogenic zone, thereby enhancing the chance of seizure freedom. Methods: Presurgical interictal MEG recordings of 94 patients (64 seizure-free > 1y post-surgery) were analyzed to extract four metrics in source space: delta power, low-to-high-frequency power ratio, functional connectivity (phase lag index), and minimum spanning tree betweenness centrality. At the group level, we estimated the overlap of the resection area with the five highest values for each metric and determined whether this overlap differed between surgery outcomes. At the individual level, those metrics were used in machine learning classifiers (linear support vector machine (SVM) and random forest) to distinguish between resection and non-resection areas and between surgery outcome groups. Results: The highest values, for all metrics, overlapped with the resection area in more than half of the patients, but the overlap did not differ between surgery outcome groups. The classifiers distinguished the resection areas from non-resection areas with 59.94% accuracy (95% confidence interval: 59.67-60.22%) for SVM and 60.34% (59.98-60.71%) for random forest, but could not differentiate seizure-free from not seizure-free patients [43.77% accuracy (42.08-45.45%) for SVM and 49.03% (47.25-50.82%) for random forest]. Significance: All four metrics localized the resection area but did not distinguish between surgery outcome groups, demonstrating that metrics derived from interictal MEG correspond to expert consensus based on several presurgical evaluation modalities, but do not yet localize the epileptogenic zone. Metrics should be improved such that they correspond to the resection area in seizure-free patients but not in patients with persistent seizures. It is important to test such localization strategies at an individual level, for example by using machine learning or individualized models, since surgery is individually tailored.

AB - Objective: Epilepsy surgery results in seizure freedom in the majority of drug-resistant patients. To improve surgery outcome we studied whether MEG metrics combined with machine learning can improve localization of the epileptogenic zone, thereby enhancing the chance of seizure freedom. Methods: Presurgical interictal MEG recordings of 94 patients (64 seizure-free > 1y post-surgery) were analyzed to extract four metrics in source space: delta power, low-to-high-frequency power ratio, functional connectivity (phase lag index), and minimum spanning tree betweenness centrality. At the group level, we estimated the overlap of the resection area with the five highest values for each metric and determined whether this overlap differed between surgery outcomes. At the individual level, those metrics were used in machine learning classifiers (linear support vector machine (SVM) and random forest) to distinguish between resection and non-resection areas and between surgery outcome groups. Results: The highest values, for all metrics, overlapped with the resection area in more than half of the patients, but the overlap did not differ between surgery outcome groups. The classifiers distinguished the resection areas from non-resection areas with 59.94% accuracy (95% confidence interval: 59.67-60.22%) for SVM and 60.34% (59.98-60.71%) for random forest, but could not differentiate seizure-free from not seizure-free patients [43.77% accuracy (42.08-45.45%) for SVM and 49.03% (47.25-50.82%) for random forest]. Significance: All four metrics localized the resection area but did not distinguish between surgery outcome groups, demonstrating that metrics derived from interictal MEG correspond to expert consensus based on several presurgical evaluation modalities, but do not yet localize the epileptogenic zone. Metrics should be improved such that they correspond to the resection area in seizure-free patients but not in patients with persistent seizures. It is important to test such localization strategies at an individual level, for example by using machine learning or individualized models, since surgery is individually tailored.

KW - magnetoencephalography

KW - presurgical evaluation

KW - functional connectivity

KW - refractory epilepsy

KW - seizure freedom

KW - beamforming

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85054932462&origin=inward

UR - https://www.ncbi.nlm.nih.gov/pubmed/30131762

U2 - 10.3389/fneur.2018.00647

DO - 10.3389/fneur.2018.00647

M3 - Article

VL - 9

JO - Frontiers in Neurology

JF - Frontiers in Neurology

SN - 1664-2295

IS - AUG

M1 - 647

ER -