Prediction error connectivity: A new method for EEG state analysis

Several models have been proposed to explain brain regional and interregional communication, the majority of them using methods that tap the frequency domain, like spectral coherence. Considering brain interareal communication as binary interactions, we describe a novel method devised to predict dyn...

Descripción completa

Detalles Bibliográficos
Autores: Principe, Alessandro, Ley Nacher, Miguel, Conesa Bertrán, Gerardo, Rocamora Zúñiga, Rodrigo Alberto
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2019
País:España
Institución:Universitat Pompeu Fabra
Repositorio:Repositorio Digital de la UPF
OAI Identifier:oai:repositori.upf.edu:10230/41822
Acceso en línea:http://hdl.handle.net/10230/41822
http://dx.doi.org/10.1016/j.neuroimage.2018.11.052
Access Level:acceso abierto
Palabra clave:Coherence
EEG states
Markov model
Seizure prediction
iEEG
id ES_8fa89f22dbfe6dcd94e965682cc79e32
oai_identifier_str oai:repositori.upf.edu:10230/41822
network_acronym_str ES
network_name_str España
repository_id_str
spelling Prediction error connectivity: A new method for EEG state analysisPrincipe, AlessandroLey Nacher, MiguelConesa Bertrán, GerardoRocamora Zúñiga, Rodrigo AlbertoCoherenceEEG statesMarkov modelSeizure predictioniEEGSeveral models have been proposed to explain brain regional and interregional communication, the majority of them using methods that tap the frequency domain, like spectral coherence. Considering brain interareal communication as binary interactions, we describe a novel method devised to predict dynamics and thus highlight abrupt changes marked by unpredictability. Based on a variable-order Markov model algorithm developed in-house for data compression, the prediction error connectivity (PEC) estimates network transitions by calculating error matrices (EMs). We analysed 20 h of EEG signals of virtual networks generated with a neural mass model. Subnetworks changed through time (2 of 5 signals), from normal to normal or pathological states. PEC was superior to spectral coherence in detecting all considered transitions, especially in broad and ripple bands. Subsequently, EMs of real data were classified using a support vector machine in order to capture the transition from interictal to preictal state and calculate seizure risk. A single seizure was randomly selected for training. Through this approach it was possible to establish a threshold that the calculated risk consistently overcame minutes before the events. Using either spectral coherence or PEC we created 1000 models that successfully predicted 6 seizures (100% sensibility), a whole cluster recorded in a patient with hippocampal epilepsy. However, PEC resulted superior to coherence in terms of true seizure free time and amount of false warnings. Indeed, the best PEC model predicted 96% of interictal time (vs. 83% of coherence) of about 20 h of stereo-EEG. This analysis was extended to patients with neo/mesocortical temporal, neocortical frontal, parietal and occipital lobe epilepsy. Again PEC showed high performance, allowing the prediction of 31 events distributed across 10 days with ROC AUCs that reached 98% (average 93 ± 5%) in 6 different patients. Moreover, considering another state transition, PEC could classify and forecast up to 88% (average 85 ± 3%) of the REM phase both in deep and scalp EEG. In conclusion, PEC is a novel approach that relies on pattern analysis in the time-domain. We believe that this method can be successfully employed both for the study of brain connectivity, and also implemented in real-life solutions for seizure detection and prediction.Elsevier20192019info:eu-repo/semantics/articleinfo:eu-repo/semantics/acceptedVersionapplication/pdfapplication/pdfhttp://hdl.handle.net/10230/41822http://dx.doi.org/10.1016/j.neuroimage.2018.11.052reponame:Repositorio Digital de la UPFinstname:Universitat Pompeu FabraInglés© Elsevier http://dx.doi.org/10.1016/j.neuroimage.2018.11.052info:eu-repo/semantics/openAccessoai:repositori.upf.edu:10230/418222026-06-12T07:21:37Z
dc.title.none.fl_str_mv Prediction error connectivity: A new method for EEG state analysis
title Prediction error connectivity: A new method for EEG state analysis
spellingShingle Prediction error connectivity: A new method for EEG state analysis
Principe, Alessandro
Coherence
EEG states
Markov model
Seizure prediction
iEEG
title_short Prediction error connectivity: A new method for EEG state analysis
title_full Prediction error connectivity: A new method for EEG state analysis
title_fullStr Prediction error connectivity: A new method for EEG state analysis
title_full_unstemmed Prediction error connectivity: A new method for EEG state analysis
title_sort Prediction error connectivity: A new method for EEG state analysis
dc.creator.none.fl_str_mv Principe, Alessandro
Ley Nacher, Miguel
Conesa Bertrán, Gerardo
Rocamora Zúñiga, Rodrigo Alberto
author Principe, Alessandro
author_facet Principe, Alessandro
Ley Nacher, Miguel
Conesa Bertrán, Gerardo
Rocamora Zúñiga, Rodrigo Alberto
author_role author
author2 Ley Nacher, Miguel
Conesa Bertrán, Gerardo
Rocamora Zúñiga, Rodrigo Alberto
author2_role author
author
author
dc.subject.none.fl_str_mv Coherence
EEG states
Markov model
Seizure prediction
iEEG
topic Coherence
EEG states
Markov model
Seizure prediction
iEEG
description Several models have been proposed to explain brain regional and interregional communication, the majority of them using methods that tap the frequency domain, like spectral coherence. Considering brain interareal communication as binary interactions, we describe a novel method devised to predict dynamics and thus highlight abrupt changes marked by unpredictability. Based on a variable-order Markov model algorithm developed in-house for data compression, the prediction error connectivity (PEC) estimates network transitions by calculating error matrices (EMs). We analysed 20 h of EEG signals of virtual networks generated with a neural mass model. Subnetworks changed through time (2 of 5 signals), from normal to normal or pathological states. PEC was superior to spectral coherence in detecting all considered transitions, especially in broad and ripple bands. Subsequently, EMs of real data were classified using a support vector machine in order to capture the transition from interictal to preictal state and calculate seizure risk. A single seizure was randomly selected for training. Through this approach it was possible to establish a threshold that the calculated risk consistently overcame minutes before the events. Using either spectral coherence or PEC we created 1000 models that successfully predicted 6 seizures (100% sensibility), a whole cluster recorded in a patient with hippocampal epilepsy. However, PEC resulted superior to coherence in terms of true seizure free time and amount of false warnings. Indeed, the best PEC model predicted 96% of interictal time (vs. 83% of coherence) of about 20 h of stereo-EEG. This analysis was extended to patients with neo/mesocortical temporal, neocortical frontal, parietal and occipital lobe epilepsy. Again PEC showed high performance, allowing the prediction of 31 events distributed across 10 days with ROC AUCs that reached 98% (average 93 ± 5%) in 6 different patients. Moreover, considering another state transition, PEC could classify and forecast up to 88% (average 85 ± 3%) of the REM phase both in deep and scalp EEG. In conclusion, PEC is a novel approach that relies on pattern analysis in the time-domain. We believe that this method can be successfully employed both for the study of brain connectivity, and also implemented in real-life solutions for seizure detection and prediction.
publishDate 2019
dc.date.none.fl_str_mv 2019
2019
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/acceptedVersion
format article
status_str acceptedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10230/41822
http://dx.doi.org/10.1016/j.neuroimage.2018.11.052
url http://hdl.handle.net/10230/41822
http://dx.doi.org/10.1016/j.neuroimage.2018.11.052
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.rights.none.fl_str_mv © Elsevier http://dx.doi.org/10.1016/j.neuroimage.2018.11.052
info:eu-repo/semantics/openAccess
rights_invalid_str_mv © Elsevier http://dx.doi.org/10.1016/j.neuroimage.2018.11.052
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:Repositorio Digital de la UPF
instname:Universitat Pompeu Fabra
instname_str Universitat Pompeu Fabra
reponame_str Repositorio Digital de la UPF
collection Repositorio Digital de la UPF
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869413230034550784
score 15,811543

Similares en LA Referencia