Trinchese, Pasquale (2021) A portable EEG-BCI framework enhanced by machine learning techniques. [Tesi di dottorato]

[img]
Preview
Text
trinchese_pasquale_33.pdf

Download (16MB) | Preview
[error in script] [error in script]
Item Type: Tesi di dottorato
Resource language: English
Title: A portable EEG-BCI framework enhanced by machine learning techniques
Creators:
CreatorsEmail
Trinchese, Pasqualepasquale.trinchese@unina.it
Date: 15 July 2021
Number of Pages: 61
Institution: Università degli Studi di Napoli Federico II
Department: Fisica
Dottorato: Fisica
Ciclo di dottorato: 33
Coordinatore del Corso di dottorato:
nomeemail
Capozziello, Salvatoresalvatore.capozziello@unina.it
Tutor:
nomeemail
Acampora, GiovanniUNSPECIFIED
Vitiello, AutiliaUNSPECIFIED
Date: 15 July 2021
Number of Pages: 61
Keywords: eeg; bci; ssvep; machine learning; artificial intelligence; brain; brain computer interface; single-channel; dry electrodes; neuroinformatics; electroencephalography; feature extraction; sensors; svm; support vector machines;
Settori scientifico-disciplinari del MIUR: Area 02 - Scienze fisiche > FIS/01 - Fisica sperimentale
Area 02 - Scienze fisiche > FIS/07 - Fisica applicata (a beni culturali, ambientali, biologia e medicina)
Area 01 - Scienze matematiche e informatiche > INF/01 - Informatica
Date Deposited: 19 Jul 2021 14:59
Last Modified: 07 Jun 2023 11:15
URI: http://www.fedoa.unina.it/id/eprint/13597

Collection description

Brain Computer Interfaces (BCIs) allow direct communication between the human brain and external devices through the processing and interpretation of brain signals. Indeed, BCI represents the ultimate achievement in human-machine interaction, eliminating all the intermediate physical steps between the intention of an action and its implementation. Electroencephalography (EEG) plays a key role in BCIs being the least invasive technique for capturing brain electrical activity. However, high performance devices turn out to be uncomfortable and of unpractical use outside dedicated facilities, mainly due to the use of many electrodes. Conversely, single-channel EEG devices made of fewer electrodes provide weak and noisy signals difficult to interpret. In this PhD thesis, a portable BCI prototype enhanced by machine learning techniques for the classification of brain signals — and in particular of Steady State Visual Evoked Potentials (SSVEPs) — is proposed. The current study embraces the design, realization, characterization, and optimization of a BCI built on top of a cost-effective single-channel EEG device. The results have been validated both in offline and online sessions thanks to the collaboration of volunteers equipped with the given prototype. Due to its usability, the proposed framework may broaden the range of state-of-the-art BCI applications.

Downloads

Downloads per month over past year

Actions (login required)

View Item View Item