Real-Time Decoding of Movement Onset and Offset for Brain-Controlled Rehabilitation Exoskeleton
Authors: Kanishka Mitra, Satyam Kumar, Frigyes Samuel Racz, Deland Liu, Ashish D. Deshpande, José del R. Millán
Organizations: Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA. · Chandra Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX, USA. · Department of Neurology, The University of Texas at Austin, Austin, TX, USA. · Walker Department of Mechanical Engineering, The University of Texas at Austin, Austin, TX, USA. · Meta Reality Labs Research, Redmond, WA, USA. · Department of Biomedical Engineering, The University of Texas at Austin, Austin, TX, USA.
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.
Figures & tables
Fig. 1 : Experimental setup, task timeline, and decoding pipeline for onset/offset EEG control of the Harmony exoskeleton. (a) Harmony (§ II-B ); (b) EEG cap; (c) stNMES pads under the cuff; (d) arm-mounted LEDs indicating trial phase (white: countdown; green: Start; red: Stop; blue: Rest); (e) fixation-based recentering (§ IV-A ); (f) subject-specific Riemannian decoders (§ II-C ); (g) three spatial targets defining goal-directed trajectories. Blue, green, and orange sliding windows depict EEG samples of shape N×C×T (trials × channels × time) with T=512 points ( 1 s at 512 Hz), as used in § II-C . Onset/offset decisions and robot transitions follow the thresholding logic in § II-C and state machine in § II-C .
Fig. 2 : Grand average spectrogram from the C3 electrode showing changes in spectral power over the course of task trials. Data were obtained by averaging over all trials and all subjects. Vertical lines mark task events (countdown, Start MI, robot moves, Stop MI, robot stops, Rest, robot returns).
Fig. 3 : Online command-delivery outcomes. Stacked bars show group-mean outcome proportions (hit, miss, timeout) for onset and offset in Sessions 2–3; numbers indicate mean hit rate. Offset is computed on attempted trials only (i.e., trials following a successful onset).
Fig. 4 : Online decoding time for hits only. Violin plots show the distribution of decision latencies for onset (cue-locked) and offset (movement-locked to assistance onset) in Sessions 2-3; black bars indicate means.
Fig. 5 : Task-based recentering shifts class separation relative to Si ( Δ median margin). Bars show the group-mean change in median margin, m=dnegative−dpositive , under task-based recentering versus the identity reference Si . Error bars denote 95% CIs across subjects (subject means over runs).
Fig. 6 : AUC by run for onset and offset under task- vs fixation-based recentering. Lines show group-mean AUC; ribbons denote 95% CIs. Inset boxes report paired Wilcoxon p -values for per-subject mean AUC differences (Task − Fix).
Chandra Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX, USA. · Department of Neurology, The University of Texas at Austin, Austin, TX, USA. · Walker Department of Mechanical Engineering, The University of Texas at Austin, Austin, TX, USA.