A composite robotic-based measure of upper limb proprioception

Abstract

Background: Proprioception is the sense of the position and movement of our limbs, and is vital for executing coordinated movements. Proprioceptive disorders are common following stroke, but clinical tests for measuring impairments in proprioception are simple ordinal scales that are unreliable and relatively crude. We developed and validated specific kinematic parameters to quantify proprioception and compared two common metrics, Euclidean and Mahalanobis distances, to combine these parameters into an overall summary score of proprioception.

Methods: We used the KINARM robotic exoskeleton to assess proprioception of the upper limb in subjects with stroke (N = 285. Mean days post-stroke = 12 ± 15). Two aspects of proprioception (position sense and kinesthetic sense) were tested using two mirror-matching tasks without vision. The tasks produced 12 parameters to quantify position sense and eight to quantify kinesthesia. The Euclidean and Mahalanobis distances of the z-scores for these parameters were computed each for position sense, kinesthetic sense, and overall proprioceptive function (average score of position and kinesthetic sense).

Results: A high proportion of stroke subjects were impaired on position matching (57%), kinesthetic matching (65%), and overall proprioception (62%). Robotic tasks were significantly correlated with clinical measures of upper extremity proprioception, motor impairment, and overall functional independence. Composite scores derived from the Euclidean distance and Mahalanobis distance showed strong content validity as they were highly correlated (r = 0.97-0.99).

Conclusions: We have outlined a composite measure of upper extremity proprioception to provide a single continuous outcome measure of proprioceptive function for use in clinical trials of rehabilitation. Multiple aspects of proprioception including sense of position, direction, speed, and amplitude of movement were incorporated into this measure. Despite similarities in the scores obtained with these two distance metrics, the Mahalanobis distance was preferred.

Keywords: Kinesthesia; Outcome measure; Position sense; Proprioception; Robotics; Stroke; Upper extremity.

Fig. 1

a KINARM robotic exoskeleton (BKIN Technologies, Kingston, ON, Canda). Subjects are seated in the wheelchair base with arms supported by the arm troughs. b Top-down view of the position matching task. The stroke affected arm was positioned by the robot (black targets, green lines) and subjects were required to mirror-match the target positions with their opposite hand (open targets, blue lines). Nine targets were matched to six times each for a total of 54 trials, presented in pseudorandom order. c Top-down view of an exemplar subject performing one trial of the kinesthetic matching task. The stroke affected arm was moved by the robot between two targets (green lines) and subjects were required to mirror match the speed, direction, and amplitude of movement as soon as they felt the robot move their arm (blue lines). The speed versus time profile represents the temporal aspects of the task, by measuring the response latency (time to initiation of the active arm movement) and peak speed ratio (difference between peak speeds of the passive (green) and active (blue) hands)

Fig. 2

Scatter plots of robotics scores for individual stroke subjects (N = 285). Greater scores indicate worse proprioception a The relationship between position matching performance calculated using subjects’ E-Scores (Euclidean distance of an individual subject’s robotic scores from the mean healthy control scores) versus M-Scores (Mahalanobis distance of an individual subject’s robotic scores from the mean healthy control scores). b Relationship between kinesthetic matching performance calculated using the E-Scores and M-Scores. c Relationship between the position matching and kinesthetic matching tasks based on the E-Scores. d Relationship between the position matching and kinesthetic matching tasks based on the M-Scores. E and M-Scores represent standard deviations from the mean of neurologically intact control performance. Grey dashed lines indicate 1.96 standard deviations. Data points beyond 1.96 indicate impaired performance. Black dotted lines on each plot indicate unity between scores, black solid lines on each plot indicate least squares fit between scores. Pearson correlation coefficients (r) and associated p-values (p) are presented in each plot

Fig. 3

Exemplar subjects’ performance on the position (left panel) and kinesthetic (middle and right panel) matching tasks. For the position matching task, the subject’s matched hand positions (open targets, blue lines) are mirrored across the vertical centre line and displayed on top of the passive robotically moved hand positions (black filled targets, green lines). For the kinesthetic matching task, both hand movements are displayed where solid green lines indicate passive robotic movements, dotted green lines indicate the optimal movement path of the opposite arm, and solid blue lines indicate active subject movements. Light blue lines indicate individual trials and dark blue lines indicate the average between all completed trials in the given movement direction. Note that for the position matching task, the blue and green lines simply connect the target positions for display purposes and do not represent the hand movements between targets. E: ‘E-Score’ indicates the subject’s composite score calculated from the Euclidean distance. M: ‘M-Score’ indicates the subject’s composite score calculated from the Mahalanobis distance. a Control exemplar. Intact position matching performance is indicated by low variability (small ellipse size), with minimal shift or contraction/expansion of the workspace (blue dotted lines). Intact kinesthetic matching performance is indicated by alignment in movement direction to the ideal movement path, and a short response latency (onset of active arm movement) with similar peak speeds between passive (green lines) and active hands (blue lines). b Stroke subject with intact performance on the position matching task. This subject also performed well on the spatial aspects of kinesthesia (middle panel) but performed poorly on the temporal aspects of kinesthesia (right panel). c Stroke subject who performed poorly on the position matching task (increased variability and shift of workspace). This subject demonstrated impairments on the spatial aspects of kinesthesia but normal performance on the temporal parameters (short and consistent response latency and peak speeds). d Stroke subject who was severely impaired on both position and kinesthetic matching tasks