Intracortical neural activity distal to seizure-onset-areas predicts human focal seizures

Timothée Proix, Mehdi Aghagolzadeh, Joseph R. Madsen, Rees Cosgrove, Emad Eskandar, Leigh R. Hochberg, Sydney S. Cash, Wilson Truccolo

Research output: Contribution to journalArticle

Abstract

The apparent unpredictability of epileptic seizures has a major impact in the quality of life of people with pharmacologically resistant seizures. Here, we present initial results and a proof-of-concept of how focal seizures can be predicted early in advance based on intracortical signals recorded from small neocortical patches away from identified seizure onset areas. We show that machine learning algorithms can discriminate between interictal and preictal periods based on multiunit activity (i.e. thresholded action potential counts) and multi-frequency band local field potentials recorded via 4 X 4 mm2 microelectrode arrays. Microelectrode arrays were implanted in 5 patients undergoing neuromonitoring for resective surgery. Post-implant analysis revealed arrays were outside the seizure onset areas. Preictal periods were defined as the 1-hour period leading to a seizure. A 5-minute gap between the preictal period and the putative seizure onset was enforced to account for potential errors in the determination of actual seizure onset times. We used extreme gradient boosting and long short-term memory networks for prediction. Prediction accuracy based on the area under the receiver operating characteristic curves reached 90% for at least one feature type in each patient. Importantly, successful prediction could be achieved based exclusively on multiunit activity. This result indicates that preictal activity in the recorded neocortical patches involved not only subthreshold postsynaptic potentials, perhaps driven by the distal seizure onset areas, but also neuronal spiking in distal recurrent neocortical networks. Beyond the commonly identified seizure onset areas, our findings point to the engagement of large-scale neuronal networks in the neural dynamics building up toward a seizure. Our initial results obtained on currently available human intracortical microelectrode array recordings warrant new studies on larger datasets, and open new perspectives for seizure prediction and control by emphasizing the contribution of multiscale neural signals in large-scale neuronal networks.

Original languageEnglish (US)
Article numbere0211847
JournalPloS one
Volume14
Issue number7
DOIs
StatePublished - Jan 1 2019
Externally publishedYes

Fingerprint

seizures
Microelectrodes
Seizures
Surgery
Learning algorithms
Frequency bands
Learning systems
prediction
Synaptic Potentials
Long-Term Memory
artificial intelligence
Short-Term Memory
ROC Curve
action potentials
Action Potentials
quality of life
neural networks
Epilepsy
Quality of Life
surgery

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Proix, T., Aghagolzadeh, M., Madsen, J. R., Cosgrove, R., Eskandar, E., Hochberg, L. R., ... Truccolo, W. (2019). Intracortical neural activity distal to seizure-onset-areas predicts human focal seizures. PloS one, 14(7), [e0211847]. https://doi.org/10.1371/journal.pone.0211847

Intracortical neural activity distal to seizure-onset-areas predicts human focal seizures. / Proix, Timothée; Aghagolzadeh, Mehdi; Madsen, Joseph R.; Cosgrove, Rees; Eskandar, Emad; Hochberg, Leigh R.; Cash, Sydney S.; Truccolo, Wilson.

In: PloS one, Vol. 14, No. 7, e0211847, 01.01.2019.

Research output: Contribution to journalArticle

Proix, T, Aghagolzadeh, M, Madsen, JR, Cosgrove, R, Eskandar, E, Hochberg, LR, Cash, SS & Truccolo, W 2019, 'Intracortical neural activity distal to seizure-onset-areas predicts human focal seizures', PloS one, vol. 14, no. 7, e0211847. https://doi.org/10.1371/journal.pone.0211847
Proix, Timothée ; Aghagolzadeh, Mehdi ; Madsen, Joseph R. ; Cosgrove, Rees ; Eskandar, Emad ; Hochberg, Leigh R. ; Cash, Sydney S. ; Truccolo, Wilson. / Intracortical neural activity distal to seizure-onset-areas predicts human focal seizures. In: PloS one. 2019 ; Vol. 14, No. 7.
@article{debc683e99174d5d8438ee841365da45,
title = "Intracortical neural activity distal to seizure-onset-areas predicts human focal seizures",
abstract = "The apparent unpredictability of epileptic seizures has a major impact in the quality of life of people with pharmacologically resistant seizures. Here, we present initial results and a proof-of-concept of how focal seizures can be predicted early in advance based on intracortical signals recorded from small neocortical patches away from identified seizure onset areas. We show that machine learning algorithms can discriminate between interictal and preictal periods based on multiunit activity (i.e. thresholded action potential counts) and multi-frequency band local field potentials recorded via 4 X 4 mm2 microelectrode arrays. Microelectrode arrays were implanted in 5 patients undergoing neuromonitoring for resective surgery. Post-implant analysis revealed arrays were outside the seizure onset areas. Preictal periods were defined as the 1-hour period leading to a seizure. A 5-minute gap between the preictal period and the putative seizure onset was enforced to account for potential errors in the determination of actual seizure onset times. We used extreme gradient boosting and long short-term memory networks for prediction. Prediction accuracy based on the area under the receiver operating characteristic curves reached 90{\%} for at least one feature type in each patient. Importantly, successful prediction could be achieved based exclusively on multiunit activity. This result indicates that preictal activity in the recorded neocortical patches involved not only subthreshold postsynaptic potentials, perhaps driven by the distal seizure onset areas, but also neuronal spiking in distal recurrent neocortical networks. Beyond the commonly identified seizure onset areas, our findings point to the engagement of large-scale neuronal networks in the neural dynamics building up toward a seizure. Our initial results obtained on currently available human intracortical microelectrode array recordings warrant new studies on larger datasets, and open new perspectives for seizure prediction and control by emphasizing the contribution of multiscale neural signals in large-scale neuronal networks.",
author = "Timoth{\'e}e Proix and Mehdi Aghagolzadeh and Madsen, {Joseph R.} and Rees Cosgrove and Emad Eskandar and Hochberg, {Leigh R.} and Cash, {Sydney S.} and Wilson Truccolo",
year = "2019",
month = "1",
day = "1",
doi = "10.1371/journal.pone.0211847",
language = "English (US)",
volume = "14",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "7",

}

TY - JOUR

T1 - Intracortical neural activity distal to seizure-onset-areas predicts human focal seizures

AU - Proix, Timothée

AU - Aghagolzadeh, Mehdi

AU - Madsen, Joseph R.

AU - Cosgrove, Rees

AU - Eskandar, Emad

AU - Hochberg, Leigh R.

AU - Cash, Sydney S.

AU - Truccolo, Wilson

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The apparent unpredictability of epileptic seizures has a major impact in the quality of life of people with pharmacologically resistant seizures. Here, we present initial results and a proof-of-concept of how focal seizures can be predicted early in advance based on intracortical signals recorded from small neocortical patches away from identified seizure onset areas. We show that machine learning algorithms can discriminate between interictal and preictal periods based on multiunit activity (i.e. thresholded action potential counts) and multi-frequency band local field potentials recorded via 4 X 4 mm2 microelectrode arrays. Microelectrode arrays were implanted in 5 patients undergoing neuromonitoring for resective surgery. Post-implant analysis revealed arrays were outside the seizure onset areas. Preictal periods were defined as the 1-hour period leading to a seizure. A 5-minute gap between the preictal period and the putative seizure onset was enforced to account for potential errors in the determination of actual seizure onset times. We used extreme gradient boosting and long short-term memory networks for prediction. Prediction accuracy based on the area under the receiver operating characteristic curves reached 90% for at least one feature type in each patient. Importantly, successful prediction could be achieved based exclusively on multiunit activity. This result indicates that preictal activity in the recorded neocortical patches involved not only subthreshold postsynaptic potentials, perhaps driven by the distal seizure onset areas, but also neuronal spiking in distal recurrent neocortical networks. Beyond the commonly identified seizure onset areas, our findings point to the engagement of large-scale neuronal networks in the neural dynamics building up toward a seizure. Our initial results obtained on currently available human intracortical microelectrode array recordings warrant new studies on larger datasets, and open new perspectives for seizure prediction and control by emphasizing the contribution of multiscale neural signals in large-scale neuronal networks.

AB - The apparent unpredictability of epileptic seizures has a major impact in the quality of life of people with pharmacologically resistant seizures. Here, we present initial results and a proof-of-concept of how focal seizures can be predicted early in advance based on intracortical signals recorded from small neocortical patches away from identified seizure onset areas. We show that machine learning algorithms can discriminate between interictal and preictal periods based on multiunit activity (i.e. thresholded action potential counts) and multi-frequency band local field potentials recorded via 4 X 4 mm2 microelectrode arrays. Microelectrode arrays were implanted in 5 patients undergoing neuromonitoring for resective surgery. Post-implant analysis revealed arrays were outside the seizure onset areas. Preictal periods were defined as the 1-hour period leading to a seizure. A 5-minute gap between the preictal period and the putative seizure onset was enforced to account for potential errors in the determination of actual seizure onset times. We used extreme gradient boosting and long short-term memory networks for prediction. Prediction accuracy based on the area under the receiver operating characteristic curves reached 90% for at least one feature type in each patient. Importantly, successful prediction could be achieved based exclusively on multiunit activity. This result indicates that preictal activity in the recorded neocortical patches involved not only subthreshold postsynaptic potentials, perhaps driven by the distal seizure onset areas, but also neuronal spiking in distal recurrent neocortical networks. Beyond the commonly identified seizure onset areas, our findings point to the engagement of large-scale neuronal networks in the neural dynamics building up toward a seizure. Our initial results obtained on currently available human intracortical microelectrode array recordings warrant new studies on larger datasets, and open new perspectives for seizure prediction and control by emphasizing the contribution of multiscale neural signals in large-scale neuronal networks.

UR - http://www.scopus.com/inward/record.url?scp=85069702550&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069702550&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0211847

DO - 10.1371/journal.pone.0211847

M3 - Article

C2 - 31329587

AN - SCOPUS:85069702550

VL - 14

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 7

M1 - e0211847

ER -