An intelligent E-skin with spatiotemporal vibrotactile cues enable personalized gait rehabilitation in Parkinson's disease

Parkinson's disease (PD), the second most prevalent neurodegenerative disorder, severely impairs gait and mobility. Pharmacologic and surgical therapies have important limitations, and existing wearable cueing devices remain clunky and poorly adaptive. To address this, we developed a novel flexible patch characterized by ultra-miniaturization and superior adaptability (FlexAdapt Patch) for gait rehabilitation in people with PD. The FlexAdapt Patch integrates a single IMU with Gait-Phase-Adaptive (GPA) algorithm and a continuously updated, personalized gait-processing strategy, delivering pre-emptive, leg-lifting vibrotactile cues. Across walking speeds, the algorithm achieved 97.79% agreement with optical motion capture as the gold standard. In a cohort of 35 PD patients, spatiotemporal vibration outperformed no stimulation and constant vibration, reducing the variance of stride length, gait-cycle duration, and support-phase ratio by 49.1–61.1%, decreasing discrepancies in hip and knee angle trajectories relative to normative curves by 38.8% and 38.2%, and lowering fatigue in four lower-limb muscles by 16.9–30.2%. In summary, the FlexAdapt Patch enables routine, home-based, personalized gait rehabilitation for PD.