Modern Brain-Computer Interface (BCI) systems have brought great technological progress, making certain tasks, specific and/or mundane, permissive even for individuals with disabling conditions to which they were previously found to be complex or difficult to perform. Such technology at present is applicable mainly in the field of communication, control of auxiliary systems for disability, and also in the field of rehabilitation, particularly in the motor field. Mainly the BCIs developed so far are divided into two macro-categories: those based on the detection of potentials evoked by external stimulation and those based on the recognition of brain activity due to the imagination of specific actions. However, there are cases where these two methodologies do not provide a good degree of control to users, particularly if the latter have particular syndromes that undermine the ability to activate certain brain traces. For this reason, other types of control signals have been analyzed, one in particular being Covert Visuospatial Attention (CVSA), which consists of the common ability to focus attention at a point in the visual field without having to move the eyes. CVSA turns out to be a good control signal because it does not require any external stimulation and can be controlled by voluntary modulation of brain signals; it is not dependent on eye movement, and thus can be used even by users with limited, or even no, voluntary gaze control; and finally it is based on the actual execution of an action instead of on imagination alone, which is beneficial because it allows the use of a signal that is intuitive and closer to everyday reality. In recent years, paradigms using CVSA as an input signal for BCI based on signals from electroencephalogram (EEG) have been developed with definitely satisfactory results, however, studies using CVSA as an online control signal for BCI have rarely been presented. This thesis aims to demonstrate that it is possible to realize such a closed-loop BCI, i.e., a system no longer based solely on data collection and an a posteriori decoding of the result, but rather in which CVSA is used precisely as an online control signal for a mental task and the user can receive, in real-time, feedback regarding that task based on the evolution of that control.
I sistemi moderni di Brain-Computer Interface (BCI) hanno portato un grande progresso tecnologico, rendendo permissivi alcuni compiti, specifici e/o mondani, anche per soggetti affetti da condizioni di disabilità a cui prima risultavano essere complessi o di difficile svolgimento. Una tale tecnologia al momento è applicabile soprattutto nel campo della comunicazione, del controllo di sistemi ausiliari per la disabilità e anche in quello della riabilitazione, in particolare nell’ambito motorio. Principalmente le BCI sviluppate finora si distinguono in due macro-categorie: quelle basate sul rilevamento di potenziali evocati da una stimolazione esterna e quelle basate sul riconoscimento dell’attività cerebrale dovuta all’immaginazione di azioni specifiche. Tuttavia ci sono dei casi in cui queste due metodologie non assicurano un buon grado di controllo agli utenti, in particolare se quest’ultimi presentano sindromi particolari che minano la capacità di attivazione di alcuni tracciati cerebrali. Per questo motivo altri tipi di segnali di controllo sono stati analizzati, uno in particolare è la Covert Visuospatial Attention (CVSA), che consiste in quella comune abilità di focalizzare l’attenzione in un punto del campo visivo senza dover muovere gli occhi. La CVSA risulta essere un buon segnale di controllo poiché non richiede alcuna stimolazione esterna e può essere controllata grazie a una modulazione volontaria dei segnali cerebrali, non dipende dal movimento oculare, e quindi può essere utilizzata anche da utenti che presentano un limitato, o anche nullo, controllo volontario dello sguardo; ed infine essa si basa sull’esecuzione effettiva di un’azione invece che sulla sola immaginazione, cosa che porta beneficio poiché permette di utilizzare un segnale intuitivo e più vicino alla realtà quotidiana. Negli ultimi anni sono stati sviluppati dei paradigmi che utilizzano la CVSA come segnale di input per BCI basate su segnali proveniente da elettroencefalogramma (EEG) con risultati decisamente soddisfacenti, tuttavia raramente sono stati presentati studi che utilizzavano la CVSA come segnale di controllo online per BCI. Questa tesi ha lo scopo di dimostrare che è possibile realizzare una tale BCI ad anello chiuso, ossia un sistema non più basato esclusivamente sulla raccolta di dati e una decodifica a posteriori del risultato, bensì nella quale la CVSA sia utilizzata proprio come segnale di controllo online per un task mentale e l’utente possa ricevere, in tempo reale, un feedback riguardante tale task in base all’evoluzione di tale controllo.
Development of a closed-loop BCI based on covert visuospatial attention
AVELLINI, RICCARDO
2023/2024
Abstract
Modern Brain-Computer Interface (BCI) systems have brought great technological progress, making certain tasks, specific and/or mundane, permissive even for individuals with disabling conditions to which they were previously found to be complex or difficult to perform. Such technology at present is applicable mainly in the field of communication, control of auxiliary systems for disability, and also in the field of rehabilitation, particularly in the motor field. Mainly the BCIs developed so far are divided into two macro-categories: those based on the detection of potentials evoked by external stimulation and those based on the recognition of brain activity due to the imagination of specific actions. However, there are cases where these two methodologies do not provide a good degree of control to users, particularly if the latter have particular syndromes that undermine the ability to activate certain brain traces. For this reason, other types of control signals have been analyzed, one in particular being Covert Visuospatial Attention (CVSA), which consists of the common ability to focus attention at a point in the visual field without having to move the eyes. CVSA turns out to be a good control signal because it does not require any external stimulation and can be controlled by voluntary modulation of brain signals; it is not dependent on eye movement, and thus can be used even by users with limited, or even no, voluntary gaze control; and finally it is based on the actual execution of an action instead of on imagination alone, which is beneficial because it allows the use of a signal that is intuitive and closer to everyday reality. In recent years, paradigms using CVSA as an input signal for BCI based on signals from electroencephalogram (EEG) have been developed with definitely satisfactory results, however, studies using CVSA as an online control signal for BCI have rarely been presented. This thesis aims to demonstrate that it is possible to realize such a closed-loop BCI, i.e., a system no longer based solely on data collection and an a posteriori decoding of the result, but rather in which CVSA is used precisely as an online control signal for a mental task and the user can receive, in real-time, feedback regarding that task based on the evolution of that control.File | Dimensione | Formato | |
---|---|---|---|
Avellini_Riccardo.pdf
accesso aperto
Dimensione
17.09 MB
Formato
Adobe PDF
|
17.09 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/80165