Show simple item record

dc.contributor.advisorGurrutxaga Goikoetxea, Ibai ORCID
dc.contributor.authorBécares Cantón, Adrián
dc.contributor.otherF. INFORMATICA
dc.contributor.otherINFORMATIKA F.
dc.date.accessioned2021-10-08T17:56:59Z
dc.date.available2021-10-08T17:56:59Z
dc.date.issued2021-10-08
dc.identifier.urihttp://hdl.handle.net/10810/53305
dc.description.abstractThe following project aims to analyze the ability of Convolutional Neural Networks(CNNs) to discriminate raw Electroencephalographic (EEG) signals for Brain-computer interfaces (BCI), in order to develop a solid and reliable model that is capable of solving these medical and clinical applications. The project also aims to serve as foundations forfuture research projects of the UPV/EHU research group Aldapa, as well as being a starting framework to apply modern techniques such as Transfer Learning or Semi-supervised Learning. To achieve this, this report collects and explains the mathematical and theoretical foundations of the architectures and models used for the development, based on the article of Schirrmeister et al. (2017) and the large EEG database provided by Kaya et al. (2018). Following the model implementation, an experimentation is designed and tested, among with an Hyperparameter Optimization setup for the developed model. Finally, the results show that the performance of the model depends on the subject and EEG recording session. It also shows that some hyperparameters influence the model, for example the optimization algorithm, but other hyperparameters barely affect the performance of the implementation.es_ES
dc.language.isoenges_ES
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectpythones_ES
dc.subjectmachine-learninges_ES
dc.subjectdeep-learninges_ES
dc.subjectEEGes_ES
dc.subjectclassificationes_ES
dc.subjectBCIes_ES
dc.subjectCNNes_ES
dc.titleBCI system for motor imagery classification using convolutional neural networkses_ES
dc.typeinfo:eu-repo/semantics/bachelorThesis
dc.date.updated2021-07-26T06:28:40Z
dc.language.rfc3066es
dc.rights.holder© 2021, el autor
dc.contributor.degreeGrado en Ingeniería Informáticaes_ES
dc.contributor.degreeInformatika Ingeniaritzako Gradua
dc.identifier.gaurregister117348-871327-11
dc.identifier.gaurassign120797-871327


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record