Motor Imagery (MI) Brain-Computer Interfaces (BCIs) represent a promising technology for neurorehabilitation and assistive control. However, the clinical viability of these systems is frequently hindered by the inherent limitations of electroencephalography (EEG) with regard to its low signal-to-noise ratio (SNR), non-stationarity, and high inter-subject variability. Standard decoding methods often fail to capture the complexity of user intention leading to unreliable performance and user frustr