Descrizione del progetto
Migliorare le interfacce cervello- computer per le persone con disabilità
Un’interfaccia cervello-computer (BCI, Brain–Computer Interface) è una tecnologia straordinaria in grado di acquisire segnali cerebrali, analizzarli e tradurli in movimento. Le BCI sono state progettate per aiutare le persone paralizzate a controllare i dispositivi ausiliari attraverso i loro pensieri. Ma a volte questa tecnologia non funziona secondo la volontà del soggetto, causando movimenti maldestri. Per risolvere questi limiti che impediscono di diffondere l’utilizzo della tecnologia, il progetto SubcorticalBCI, finanziato dall’UE, svilupperà un nuovo tipo di BCI che legge l’attività del corpo striato, ovvero la parte del cervello associata alla motricità fine. La nuova BCI sarà inoltre più stabile e semplice da usare.
Obiettivo
Every year, half a million people become paralysed by a spinal cord injury. Assistive technologies, such as Brain-Computer Interfaces (BCIs), can improve their mobility, independence, and overall well-being. BCIs bypass a neurological injury by reading the user’s intent from their brain activity and using it to control a computer cursor or a robotic arm, or even to reanimate their own paralysed limbs. Despite these remarkable feats, BCIs still face challenges that prevent their widespread use.
Here, I propose to address two of their main shortcomings: their unintuitive control that produces unskilled movements, and their instable performance that decays over time. The brain controls movement by coordinating the activity of many areas. Thus, unintuitive BCI control might be due to their reliance on the activity of a single cortical region (typically motor cortex) to decode the user’s intent. I will develop a new type of BCI that reads the activity of the striatum, a subcortical area that receives inputs from the entire cortex and has been shown to be critical for skilled movements. My hypothesis is that a BCI based on striatal activity will mimic the execution of a skilled movement. Using large-scale neural recording techniques and emerging computational techniques, I will identify striatal population dynamics and use them as input for the BCI, expecting to show that such an approach outperforms current BCIs. Next, I will address the instability problem. BCI instability is mainly caused by the inevitable changes in recorded neurons over long timescales. My host has recently developed a method that reveals the ‘true’ cortical dynamics underlying a given behaviour. I will adopt this method for the proposed BCI to stabilise its performance over long time periods.
If successful, this project will lead to BCIs that are easier to use, more precise, and stable. Their future translation to humans could bring BCIs closer to the clinic, with considerable socioeconomic impact.
Parole chiave
Programma(i)
Argomento(i)
Invito a presentare proposte
(si apre in una nuova finestra) H2020-MSCA-IF-2020
Vedi altri progetti per questo bandoMeccanismo di finanziamento
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinatore
SW7 2AZ LONDON
Regno Unito