Deep learning and brain-computer interfaces

Author(s)

Willian Rampazzo

• Document type: Master's Dissertation
• Press: Campinas, SP.
• Institution: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Elétrica e de Computação
• Defense date: 2019-08-13

Examining board members

Romis Ribeiro de Faissol Attux; André Kazuo Takahata; Fernando José Von Zuben

• Advisor: Romis Ribeiro de Faissol Attux

Abstract

A Brain-Computer Interface (BCI) is an artificial system of direct communication between the brain and an external device. It operates independently of peripheral nerves and muscles. Its purpose is to translate the user's intention, associated with the measurement of brain-originated activity, into control signals corresponding to those of the application. To do so, it makes use of signal processing and pattern recognition techniques. Different areas can benefit from BCIs, such as the entertainment area and the healthcare area. Especially in healthcare, these systems can have a significant impact on people with diseases that lead to locked-in syndrome and are possibly the only option for communication in such cases. Deep neural networks have been applied to problems in different areas, such as computer vision, natural language processing, medical diagnostics, among others, obtaining results that often surpass the state-of-the-art. An exciting feature of these networks, in particular, Deep Convolutional Neural Networks (DCNNs), is the ability to extract features automatically, often without the need for manual attribute engineering, an almost mandatory step for a large number of machine learning techniques. The promising results obtained by DCNNs in different areas indicate that there is potential to employ them in the task of processing brain signals in BCI systems. In this work, we explore the use of these networks in a Steady-State Visually Evoked Potentials (SSVEP) -based BCI investigating, initially, whether different input formats such as the raw signal or a signal transformation, for example, the short-time Fourier transform, influence the performance of DCNNs. For such, we propose new DCNN architectures and evaluate them in different input formats, comparing their performances to the approaches usually adopted in the BCI classification step. This work also investigates whether it is possible to use the knowledge transfer technique to adjust pre-trained DCNNs to BCI data. Experimental results indicate that DCNNs are an option to apply to brain signal processing in BCIs (AU)

• FAPESP's process: 18/04100-3 - Deep learning and brain-computer interfaces
• Grantee: Willian Rampazzo
• Support Opportunities: Scholarships in Brazil - Master