Skip to main content
Fig. 1 | Journal of NeuroEngineering and Rehabilitation

Fig. 1

From: The use of machine learning and deep learning techniques to assess proprioceptive impairments of the upper limb after stroke

Fig. 1

Arm Position Matching (APM) task. A The Kinarm exoskeleton robot. B Typical healthy control participant data. The robot moved the participant’s passive right hand to one of 9 spatial locations (filled symbols). The participant then attempted to mirror match with the left active hand (open symbols). The solid blue line connects the average final positions of the outer eight target locations of the matching hand (active hand). Solid green line connects the outer eight targets for the robot moved passive hand. The dashed blue line is the mirror reflection of solid blue line, which allows a visual comparison of the average final outer 8 positions of the active and passive (robot-moved) hands. Ellipses represent one standard deviation of the matched positions. The ellipses represent trial-to-trial variability, where a larger ellipse means the participant was less consistent (i.e., more variable) in matching the position of their passive hand with the active hand. C An exemplar stroke participant who demonstrated high variability in position matching. D An exemplar stroke participant who demonstrated a contracted sense of their workspace. E An exemplar stroke participant who demonstrated a spatial shift of their workspace

Back to article page