Real-Time Decoding of Movement Onset and Offset for Brain-Controlled Rehabilitation Exoskeleton

2026-03-17

AI Summary

This study addresses the challenge of aligning user movement intent with neuroplasticity requirements in upper-limb rehabilitation exoskeletons, which traditionally rely on limb signals and struggle with precise synchronization. The authors propose an online, dual-state control framework based on non-invasive electroencephalography (EEG) that enables users to voluntarily initiate and terminate rehabilitative movements. Key innovations include a category-agnostic gaze recalibration strategy that mitigates signal drift without introducing class bias, and a threshold-free control scheme integrating online decoding, dual-state (initiation/termination) classification, and asymmetric boundary diagnostics for high discriminability. Experiments with eight participants demonstrate initiation and termination hit rates of 61.5% and 64.5%, respectively, alongside significant AUC improvements (56% for initiation, 34% for termination), effectively alleviating intra- and inter-day performance degradation.

Technology Category

• Intelligent Robots: State Estimation
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Humans and AI: Brain-Sensing and Analysis

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

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.

Problem

• brain-computer interface
• motor imagery
• rehabilitation exoskeleton
• movement onset
• movement offset

Innovation

• real-time EEG decoding
• dual-state motor imagery
• class-agnostic recentering
• brain-controlled exoskeleton
• neuroplasticity-targeted rehabilitation