REHAB: A wearable sensor–based kinematic dataset collected under standardized rehabilitation tasks from 120 stroke patients

REHAB is a wearable sensor–based kinematic dataset collected from 120 post-stroke patients during a three-week rehabilitation program. The dataset includes inertial measurement unit (IMU) and flex sensor recordings from 27 standardized rehabilitation assessment movements and 16 rehabilitation training movements, accompanied by corresponding clinical scores and movement labels. Designed to support data-driven stroke rehabilitation research, REHAB-120 provides a comprehensive and well-annotated resource for motor function assessment, rehabilitation training analysis, and action recognition.