A User-Driven Wearable Hand Robot for Functional Task Training After Stroke

Abstract

Date Presented 03/27/20

Regaining adequate hand function for participation is a challenge for many stroke survivors. Eleven participants completed 12 grasp, transport, and release training sessions with a user-driven hand robot that provides active-assist for finger extension. Mean gain of 2.63 points (P = .03) on the Fugl-Meyer was observed, with most gains in the distal subscale (2.27 points, P < .001). This intervention is a promising adjunct to conventional therapy to promote typical grasp patterns and functional use.

Primary Author and Speaker: Sangwoo Park

Additional Authors and Speakers: Daniel Geller

Contributing Authors: Sangwoo Park, Lynne Weber, Cassie Meeker, Lauri Bishop, Joel Stein, Matei Ciocarlie

PURPOSE: Upper extremity (UE) hemiparesis is a challenging complication of stroke. Regaining sufficient hand function for independent participation in occupations remains a challenge. Robotics as an intervention tool for chronic stroke survivors can facilitate remediation or serve as a compensatory device for participation. The robot in this study is a wearable, user-driven, active-assist device that provides powered assistance for finger extension while allowing active flexion.

DESIGN: This interventional pilot case series used a purposive sample and a pre-test, post-test design. For inclusion, adults had to consent; be at least 6 months post stroke; have passive range at the wrist to neutral and digits within functional limits; have active range to 30° shoulder flexion, 20° shoulder abduction, and 20° elbow flexion; have gross grasp; and have at least trace digit extension. Participants were excluded if they had a comorbidity or recent botulinum toxin injection (<13 weeks) in the affected UE, had another neurological disorder, or had significant spasticity (>2 on Modified Ashworth Scale).

METHOD: 11 participants completed 12 training sessions over 4 weeks. Each 30-minute session involved repetition of grasp, transport, and release tasks with a variety of common objects, bimanual tasks, and simulated ADLs.

Outcome measures were the Fugl-Meyer UE (FM), Action Research Arm Test (ARAT), and Box and Block Test (BBT). Assessments were performed once at baseline and twice after intervention: all without the device and ARAT and BBT with robotic assistance.

The device has a motor mounted on a dorsal forearm splint that powers 4 cables to act as ’tendons’ for digits 2-5. Fingertip mounts increase the tendons’ moment arms over joints to allow the motor to overcome spasticity without exerting high forces. The thumb is splinted in opposition. The device can use two control mechanisms to infer the participant’s intention to open, relax, or close the hand. The first intuitive but inconsistent method uses an ipsilateral EMG forearm band to detect pattern recognition of signals. The second method, which uses an instrumented contralateral shoulder harness, is more robust, but requires learning a new motor pattern (e.g. shoulder depression triggers robot assisted hand opening).

RESULTS: Analysis with paired sample t-tests demonstrated statistically significant mean gains on the FM of 2.63 (P=.03). Notably, most gains occurred in the distal (hand/wrist) subscale (2.27 points, P<.001). The device shows promise as a restorative tool.

Statistically significant gains were not achieved globally on ARAT and BBT, but positive trends were noted. Lower functioning subjects had a mean gain of 2.4 (P=.03) on the grasp subtest of ARAT with robotic assistance. Those who scored zero (non-functional) on BBT at baseline had a mean gain of 2 (P=.014) with robotic assistance. This suggests utility as an assistive device for clients with limited capacity.

Critically, participants reported enjoyment, functional improvement, and a desire to continue training. They stated: “It encourages me to use my hand more“; “I can mop with 2 hands and do my own toothpaste”; “If I could, I would wear it for everything.”

CONCLUSION: Training with this device is a promising intervention for stroke survivors as an adjunct to conventional therapy in order to promote typical grasp patterns and increased functional use of the hand. Further study is warranted to respond to user feedback and address limitations.

IMPACT STATEMENT: Grasp, transport, and release training with this robot device yielded functional gains and was acceptable to clients. For chronic stroke survivors, wearable robotics may facilitate functional recovery and encourage use of their hemiparetic UE in daily occupations.

References

Schneider, E. J., Lannin, N. A., Ada, L., & Schmidt, J. (2016). Increasing the amount of usual rehabilitation improves activity after stroke: a systematic review. Journal of Physiotherapy, 62(4). doi:https://doi.org/10.1016/j.jphys.2016.08.006

French B, Thomas LH, Coupe J, McMahon NE, Connell L, Harrison J, Sutton CJ, Tishkovskaya S, Watkins CL. Repetitive task training for improving functional ability after stroke. Cochrane Database of Systematic Reviews, 11. doi:10.1002/14651858.CD006073.pub3.

Park, S., Weber, L., Bishop, L., Stein, J., & Ciocarlie, M. (2018). Design and development of effective transmission mechanisms on a tendon driven hand orthosis for stroke patients. Robotics and Automation (ICRA), 2018. doi:10.1109/ICRA.2018.8461069