A deep learning approach to artifact removal in Transcranial Electrical Stimulation: from shallow methods to deep neural networks and state space models

Vista sencilla de ítem
dc.contributor.authorFernandez De Retana Uribe, Miguel
dc.contributor.authorMatanzas de Luis, Pablo
dc.contributor.authorPeña Lasa, Javier
dc.contributor.authorAlmeida, Aitor
dc.date.accessioned2025-11-07T09:33:24Z
dc.date.available2025-11-07T09:33:24Z
dc.date.issued2025-11-19
dc.date.updated2025-11-07T09:33:24Z
dc.description.abstractTranscranial Electrical Stimulation (tES) is a non-invasive neuromodulation technique that generates artifacts in simultaneous EEG recordings, hindering brain activity analysis. This study analyzes Machine Learning (ML) methods for tES noise artifact removal across three stimulation types: tDCS, tACS, and tRNS. Synthetic datasets were created by combining clean EEG data with synthetic tES artifacts. Eleven artifact removal techniques were tested and evaluated using the Root Relative Mean Squared Error (RRMSE) in the temporal and spectral domains, and the Correlation Coefficient (CC). Results indicate that method performance is highly dependent on stimulation type: for tDCS, a convolutional network (Complex CNN) performed best; while a multi-modular network (M4) based on State Space Models (SSMs) yielded the best results for tACS and tRNS. This study provides guidelines for selecting efficient artifact removal methods for different tES modalities, establishing a benchmark for future research in this area and paving the way for more robust analysis of neural dynamics in advanced clinical and neuroimaging applications.en
dc.description.sponsorshipThis work has been supported by DEUSTEK5 – Human-Centric Com-puting for Smart Sustainable Communities and Environments, Basque Universities’ System’s research group, with code IT1582-22en
dc.identifier.citationFernandez-de-Retana, M., Matanzas-de-Luis, P., Peña, J., & Almeida, A. (2025). A deep learning approach to artifact removal in Transcranial Electrical Stimulation: from shallow methods to deep neural networks and state space models. Neuroscience, 588, 152-159. https://doi.org/10.1016/J.NEUROSCIENCE.2025.10.004
dc.identifier.doi10.1016/J.NEUROSCIENCE.2025.10.004
dc.identifier.eissn1873-7544
dc.identifier.issn0306-4522
dc.identifier.urihttps://hdl.handle.net/20.500.14454/4313
dc.language.isoeng
dc.publisherElsevier Ltd
dc.rights© 2025 The Authors
dc.subject.otherBioinformatics
dc.subject.otherDeep learning (DL)
dc.subject.otherEEG denoising
dc.subject.otherElectroencephalogram (EEG)
dc.subject.otherNoise filtering
dc.subject.otherState space models (SSM)
dc.subject.otherTranscranial Electrical Stimulation (tES)
dc.titleA deep learning approach to artifact removal in Transcranial Electrical Stimulation: from shallow methods to deep neural networks and state space modelsen
dc.typejournal article
dcterms.accessRightsopen access
oaire.citation.endPage159
oaire.citation.startPage152
oaire.citation.titleNeuroscience
oaire.citation.volume588
oaire.licenseConditionhttps://creativecommons.org/licenses/by/4.0/
oaire.versionVoR
Archivos
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
fernandez_deep_2025.pdf
Tamaño:
1.07 MB
Formato:
Adobe Portable Document Format
Descargar
Colecciones
Artículos