Institute of Measurement Science
Topic
Towards an ecologically valid BCI and head-mounted VR system for post-stroke neurorehabilitation
PhD. program
Measurement technology
Year of admission
2024
Name of the supervisor
Ing. Mgr. Roman Rosipal, DrSc.
Contact:
Receiving school
Faculty of Electrical Engineering and Informatics at the Slovak University of Technology in Bratislava
Annotation
A growing body of evidence suggests that intelligent systems based on brain-computer interfaces (BCI) provide a flexible platform for a series of neurorehabilitation therapies. Additionally, the recent development of augmented or virtual reality (VR) technologies moved BCI forward by creating flexible environments where modern human-computer interactions are necessary. Head-mounted VR displays (HMD) move standard 2D screen-based VR into a virtual 3D world. This shift brings another level of flexibility and enables the creation of immersive virtual worlds.
The dissertation is focused on the area of motor neurorehabilitation in patients after stroke. Rehabilitation takes place using the BCI-HMD system, which consists of BCI and feedback in the form of visualization in a VR environment. The dissertation's practical part aims to improve the software and hardware part of the existing system to create a robust, compact, easily applicable, and user-acceptable system. This section focuses on measuring, evaluating, and analyzing electrical brain activity through scalp-recorded electroencephalography (EEG). The second part of the dissertation is focused on developing advanced machine learning (ML) tools and algorithms for detecting EEG changes associated with mental imagery of movement and changes in subjects' mental and cognitive states during BCI training. Attention will be paid to understanding the brain processes related to activities in the BCI-HMD environment. This goal will be achieved by carefully balancing existing neurophysiological evidence and knowledge with the information provided by ML tools that process recorded brain activity, and it will be verified through detailed experimental testing.
The dissertation is focused on the area of motor neurorehabilitation in patients after stroke. Rehabilitation takes place using the BCI-HMD system, which consists of BCI and feedback in the form of visualization in a VR environment. The dissertation's practical part aims to improve the software and hardware part of the existing system to create a robust, compact, easily applicable, and user-acceptable system. This section focuses on measuring, evaluating, and analyzing electrical brain activity through scalp-recorded electroencephalography (EEG). The second part of the dissertation is focused on developing advanced machine learning (ML) tools and algorithms for detecting EEG changes associated with mental imagery of movement and changes in subjects' mental and cognitive states during BCI training. Attention will be paid to understanding the brain processes related to activities in the BCI-HMD environment. This goal will be achieved by carefully balancing existing neurophysiological evidence and knowledge with the information provided by ML tools that process recorded brain activity, and it will be verified through detailed experimental testing.