CURRENT (AND PREVIOUS) FUNCTION: Arnau DIllen is a PhD student at the MFYS research group of the Vrije Universiteit Brussel. In 2018 he received his masters degree in computer science with a specialization in artificial intelligence. During the period between receiving his masters degree and starting his PhD, he worked as a software engineer for the Inter University Institute for Bioinformatics in Brussels.
RESEARCH AREA: Arnau's research is about developing Brain Computer Interfaces (BCI). These allow a user to control a device using physiological signals related to neural activity. The translation of these signals to a command is typically achieved with artificial intelligence techniques. The final goal of this project is to create a BCI that is still performant when using commercially available devices to measure the signals. The signal with these commercial devices is more noisy than with medical devices, which are only available in hospitals or research labs.
CURRENT RESEARCH QUESTIONS: To deal with the noisy signals from commercial devices, performant artificial intelligence methods are required. The research question for Arnau is to investigate if Deep Learning would be a good method for BCI. Such models allow for the raw signal to be translated directly to a command from the user. The next question would be wether these model would still work well in an online setting where signals have to be decoded in real time.
PRACTICAL APPLICABILITY: The final goal of Arnau's research would be to integrate the final model in a real-world application that would allow a user to control a device with their mind. Considered devices include exoskeletons, robotic prosthesis and control in a virtual environment among others. A successful conclusion to the project would bring us one step closer to commercial availability of BCI for control of everyday devices.