https://www.mdu.se/

Practicing Motor Imagery (MI) with a Brain-Computer Interface (BCI) has shown promise in promoting motor recovery in stroke patients. A BCI records a person’s brain activity and provides feedback to the person in real time, which allows the person to practice his or her brain activity. By imagining a movement (performing MI) such as gripping with their hand, cortical areas in the brain are activated that largely overlaps with those activated during the actual hand movement. A BCI can provide positive feedback when the hand-related cortical areas are activated during MI, which helps a person to learn how to perform MI. Despite evidence that stroke patients may recover some motor function from practicing MI with BCI feedback thanks to the feedback provided from a BCI, the effectiveness and reliability of BCI-based rehabilitation are still poor. 

A BCI can detect MI by analyzing patterns of features from the brain activity. The most common features are extracted from the oscillatory activity in the brain.  In BCI research, MI is often treated as a static pattern of features, which is detected by using machine learning algorithms to assign activity into a binary state. However, this model of MI may be inaccurate. Analyzing brain activity as dynamically varying over time and with a continuous measure of strength could better represent the cortical activity related to MI. 

In this Licentiate thesis, I explore a method for analyzing the temporal dynamic of MI-activity with a continuous measure of strength. Brain activity was recorded with electroencephalography (EEG) and subject-specific feature patterns were extracted from a group of healthy subjects while they performed MI of two opposing hand movements: opening and closing the hand. Although MI of the two same-hand movements could not be discriminated, the continuous output from a machine learning algorithm was shown to correlate well with MI-related feature patterns. The temporal analysis also revealed that MI is dynamically encoded early, but later stabilizes into a more static pattern of brain activity. Last, to accommodate for higher temporal resolution of MI, I designed and evaluated a BCI framework by its feedback delay and uncertainty as a function of the stress on the system and found a non-linear correlation. These results could be essential for developing a BCI with time-critical feedback.

To summarize, in this Licentiate thesis I propose a promising method for analyzing and extracting a temporal representation of MI, enabling relevant and continuous neurofeedback which may contribute to clinical advances in BCI-based stroke rehabilitation.

Stroke is a major cause of long-term disability, often resulting in severe motor impairments. Motor imagery has been proposed as a technique to engage motor networks and promote neuroplasticity, even in the absence of physical movement. Brain–computer interfaces (BCIs) can decode neural activity and provide feedback in a closed loop, supporting reinforcement-based motor learning. BCIs have shown modest but promising results, and their efficiency and usability can be improved by understanding how design choices influence human performance and clinical outcomes. Key challenges include achieving high signal precision through high-dimensional data, delivering real-time feedback under strict latency constraints, and managing non-stationary brain signals within and across sessions. This thesis presents an iterative and exploratory investigation of these challenges. Methods were first developed and validated in healthy participants and later implemented in a clinical intervention with stroke patients. Contributions include classification of motor imagery tasks involving one hand, characterization of motor signals in high-dimensional data, evaluation of system latency for real-time feedback, application of multivariate pattern analysis to capture temporal dynamics of the motor signal, introduction of a continuous metric for motor signal strength linked to visual representations, implementation of a method to set and maintain difficulty level during BCI training, evaluation of adaptive classification methods to maintain performance across sessions, and validation of these methods in a stroke case study. The findings advance understanding of how BCI design parameters affect usability and clinical relevance. By integrating neuroscience, machine learning, and principles of motor learning and neuroplasticity, this work contributes to the development of more effective and personalized BCI systems for stroke rehabilitation.