Enhancing Stroke Rehabilitation with Non-invasive BCI and embodied VR

Stroke is a major cause of long-term disability, with cases increasing annually and affecting not only older adults but also a growing number of younger individuals. It leads to severe cognitive and motor impairments, reducing independence and quality of life. Additionally, stroke often results in emotional and psychological challenges, including depression and anxiety, further complicating recovery. The long-term need for rehabilitation places a heavy burden on both patients and healthcare systems, requiring innovative solutions to improve recovery outcomes efficiently and affordably.

One promising approach involves the use of non-invasive Brain-Computer Interfaces (BCIs) that leverage motor imagery (MI) to directly engage and retrain the nervous system. By translating brain activity into actionable commands, BCIs offer an alternative pathway for motor recovery, even for patients with limited physical movement. However, despite their potential, current BCI-based rehabilitation methods face challenges, particularly in skill acquisition and maintaining long-term effectiveness. Many stroke patients struggle with accurately controlling BCIs, limiting their ability to fully benefit from these interventions. Virtual Reality (VR) has emerged as a valuable tool in rehabilitation by providing an immersive, controlled environment for motor and cognitive training. VR-based therapy enhances engagement, motivation, and adherence to training while offering real-time feedback and adaptive difficulty levels, which are essential for personalized rehabilitation programs. The combination of VR and BCI has the potential to create a more effective and engaging rehabilitation approach.

This project seeks to overcome these limitations by developing a more inclusive and adaptive rehabilitation system. By integrating advanced neuroimaging techniques such as EEG and fMRI, we aim to better understand neural activity patterns associated with motor imagery. Additionally, machine learning algorithms will be employed to enhance BCI accuracy, improving responsiveness and adaptability for individual users. Finally, neuroadaptive VR will be incorporated to create an engaging, immersive training environment that dynamically adjusts to each patient's progress, maximizing rehabilitation outcomes. Through this multi-faceted approach, we aim to improve motor skill acquisition, increase patient engagement, and ultimately enhance the effectiveness of stroke rehabilitation.