Cargando…

Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG

Seizure detection algorithms are often optimized to detect seizures from the epileptogenic cortex. However, in non-localizable epilepsies, the thalamus is frequently targeted for neuromodulation. Developing a reliable seizure detection algorithm from thalamic SEEG may facilitate the translation of c...

Descripción completa

Detalles Bibliográficos
Autores principales: Ganti, Bhargava, Chaitanya, Ganne, Balamurugan, Ridhanya Sree, Nagaraj, Nithin, Balasubramanian, Karthi, Pati, Sandipan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8889931/
https://www.ncbi.nlm.nih.gov/pubmed/35250803
http://dx.doi.org/10.3389/fneur.2022.755094
_version_ 1784661518799863808
author Ganti, Bhargava
Chaitanya, Ganne
Balamurugan, Ridhanya Sree
Nagaraj, Nithin
Balasubramanian, Karthi
Pati, Sandipan
author_facet Ganti, Bhargava
Chaitanya, Ganne
Balamurugan, Ridhanya Sree
Nagaraj, Nithin
Balasubramanian, Karthi
Pati, Sandipan
author_sort Ganti, Bhargava
collection PubMed
description Seizure detection algorithms are often optimized to detect seizures from the epileptogenic cortex. However, in non-localizable epilepsies, the thalamus is frequently targeted for neuromodulation. Developing a reliable seizure detection algorithm from thalamic SEEG may facilitate the translation of closed-loop neuromodulation. Deep learning algorithms promise reliable seizure detectors, but the major impediment is the lack of larger samples of curated ictal thalamic SEEG needed for training classifiers. We aimed to investigate if synthetic data generated by temporal Generative Adversarial Networks (TGAN) can inflate the sample size to improve the performance of a deep learning classifier of ictal and interictal states from limited samples of thalamic SEEG. Thalamic SEEG from 13 patients (84 seizures) was obtained during stereo EEG evaluation for epilepsy surgery. Overall, TGAN generated synthetic data augmented the performance of the bidirectional Long-Short Term Memory (BiLSTM) performance in classifying thalamic ictal and baseline states. Adding synthetic data improved the accuracy of the detection model by 18.5%. Importantly, this approach can be applied to classify electrographic seizure onset patterns or develop patient-specific seizure detectors from implanted neuromodulation devices.
format Online
Article
Text
id pubmed-8889931
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-88899312022-03-03 Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG Ganti, Bhargava Chaitanya, Ganne Balamurugan, Ridhanya Sree Nagaraj, Nithin Balasubramanian, Karthi Pati, Sandipan Front Neurol Neurology Seizure detection algorithms are often optimized to detect seizures from the epileptogenic cortex. However, in non-localizable epilepsies, the thalamus is frequently targeted for neuromodulation. Developing a reliable seizure detection algorithm from thalamic SEEG may facilitate the translation of closed-loop neuromodulation. Deep learning algorithms promise reliable seizure detectors, but the major impediment is the lack of larger samples of curated ictal thalamic SEEG needed for training classifiers. We aimed to investigate if synthetic data generated by temporal Generative Adversarial Networks (TGAN) can inflate the sample size to improve the performance of a deep learning classifier of ictal and interictal states from limited samples of thalamic SEEG. Thalamic SEEG from 13 patients (84 seizures) was obtained during stereo EEG evaluation for epilepsy surgery. Overall, TGAN generated synthetic data augmented the performance of the bidirectional Long-Short Term Memory (BiLSTM) performance in classifying thalamic ictal and baseline states. Adding synthetic data improved the accuracy of the detection model by 18.5%. Importantly, this approach can be applied to classify electrographic seizure onset patterns or develop patient-specific seizure detectors from implanted neuromodulation devices. Frontiers Media S.A. 2022-02-16 /pmc/articles/PMC8889931/ /pubmed/35250803 http://dx.doi.org/10.3389/fneur.2022.755094 Text en Copyright © 2022 Ganti, Chaitanya, Balamurugan, Nagaraj, Balasubramanian and Pati. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neurology
Ganti, Bhargava
Chaitanya, Ganne
Balamurugan, Ridhanya Sree
Nagaraj, Nithin
Balasubramanian, Karthi
Pati, Sandipan
Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG
title Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG
title_full Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG
title_fullStr Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG
title_full_unstemmed Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG
title_short Time-Series Generative Adversarial Network Approach of Deep Learning Improves Seizure Detection From the Human Thalamic SEEG
title_sort time-series generative adversarial network approach of deep learning improves seizure detection from the human thalamic seeg
topic Neurology
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8889931/
https://www.ncbi.nlm.nih.gov/pubmed/35250803
http://dx.doi.org/10.3389/fneur.2022.755094
work_keys_str_mv AT gantibhargava timeseriesgenerativeadversarialnetworkapproachofdeeplearningimprovesseizuredetectionfromthehumanthalamicseeg
AT chaitanyaganne timeseriesgenerativeadversarialnetworkapproachofdeeplearningimprovesseizuredetectionfromthehumanthalamicseeg
AT balamuruganridhanyasree timeseriesgenerativeadversarialnetworkapproachofdeeplearningimprovesseizuredetectionfromthehumanthalamicseeg
AT nagarajnithin timeseriesgenerativeadversarialnetworkapproachofdeeplearningimprovesseizuredetectionfromthehumanthalamicseeg
AT balasubramaniankarthi timeseriesgenerativeadversarialnetworkapproachofdeeplearningimprovesseizuredetectionfromthehumanthalamicseeg
AT patisandipan timeseriesgenerativeadversarialnetworkapproachofdeeplearningimprovesseizuredetectionfromthehumanthalamicseeg