Robotic task specific training for upper limb neurorehabilitation: a mixed methods feasibility trial reporting achievable dose

Vincent Crocher,Kim Brock,Janine Simondson,Marlena Klaic,Mary P Galea

DOI: https://doi.org/10.1080/09638288.2024.2394175

2024-08-27

Abstract:Purpose: Robotic devices for upper-limb neurorehabilitation allow an increase in intensity of practice, often relying on video game-based training strategies with limited capacity to individualise training and integrate functional training. This study shows the development of a robotic Task Specific Training (TST) protocol and evaluate the achieved dose. Materials and methods: Mixed-method study. A 3D robotic device for the upper limb, was made available to therapists for use during neurorehabilitation sessions. A first phase allowed clinicians to define a dedicated session protocol for TST. In a second phase the protocol was applied and the achieved dose was measured. Results: First phase (N = 5): a specific protocol, using deweighting for assessment, followed by customised passive movements and then active movement practice was developed. Second phase: the protocol was successfully applied with all participants (N = 10). Intervention duration: 4.5 ± 0.8 weeks, session frequency: 1.4 ± 0.2sessions/week, session length: 42 ± 9mins, session density: 39 ± 13%, intensity: 214 ± 84 movements/session, difficulty: dn = 0.77 ± 0.1 (normalised reaching distance) and Ɵ = 6.3 ± 23° (transverse reaching angle). Sessions' density and intensity were consistent across participants but clear differences of difficulty were observed. No changes in metrics were observed over the intervention. Conclusions: Robotic systems can support TST with high therapy intensity by modulating the practice difficulty to participants' needs and capabilities.