Search papers, labs, and topics across Lattice.
This paper presents a non-invasive EEG-based brain-computer interface (BCI) system for real-time control of an upper-limb exoskeleton, enabling both initiation and termination of movement based on motor imagery. They identify and address a class-driven bias in task-based recentering methods by introducing a class-agnostic fixation-based recentering approach, significantly improving decoding accuracy for movement onset and offset. The system achieves group-mean hit rates of 61.5% for onset and 64.5% for offset in online sessions, demonstrating reliable start-stop control.
Achieve intention-driven start-stop control of a rehabilitation exoskeleton from non-invasive EEG by fixing a common bias in task-based recentering.
Robot-assisted therapy can deliver high-dose, task-specific training after neurologic injury, but most systems act primarily at the limb level-engaging the impaired neural circuits only indirectly-which remains a key barrier to truly contingent, neuroplasticity-targeted rehabilitation. We address this gap by implementing online, dual-state motor imagery control of an upper-limb exoskeleton, enabling goal-directed reaches to be both initiated and terminated directly from non-invasive EEG. Eight participants used EEG to initiate assistance and then volitionally halt the robot mid-trajectory. Across two online sessions, group-mean hit rates were 61.5% for onset and 64.5% for offset, demonstrating reliable start-stop command delivery despite instrumental noise and passive arm motion. Methodologically, we reveal a systematic, class-driven bias induced by common task-based recentering using an asymmetric margin diagnostic, and we introduce a class-agnostic fixation-based recentering method that tracks drift without sampling command classes while preserving class geometry. This substantially improves threshold-free separability (AUC gains: onset +56%, p = 0.0117; offset +34%, p = 0.0251) and reduces bias within and across days. Together, these results help bridge offline decoding and practical, intention-driven start-stop control of a rehabilitation exoskeleton, enabling precisely timed, contingent assistance aligned with neuroplasticity goals while supporting future clinical translation.
