Skip to main content

22-08-2011 | Physical rehabilitation | Article

ADAPT robotic system shows promise for improving stroke rehabilitation


Free abstract

MedWire News: The Adaptive and Automatic Presentation of Tasks (ADAPT) robotic practice system shows promise for helping patients to improve hand and arm function after experiencing stroke, say researchers.

Previous research suggests that intensive task-specific practice can improve hand and arm function after stroke, but current evidence indicates that such patients rarely undergo the optimum amount of physical therapy needed for these tasks to be effective.

Nicolas Schweighofer (University of Southern California, Los Angeles, USA) and colleagues explain that to try and increase the amount of physical practice carried out by stroke patients, robotic systems such as ADAPT have been developed to add to the treatment provided by human therapists.

In this study, Schweighofer and team tested the efficacy of the ADAPT system in five patients with chronic stroke and mild-to-moderate upper body function impairment.

The patients were aged 66.2 years on average, were all more than 9 months post-stroke, and had a mean Fugl-Meyer arm score of 49.2 (90° shoulder flexion and 30° elbow extension).

At each trial the ADAPT robotic system selects a functional task for each patient according to a training schedule and based on that individual's previous performance. The tasks include activities such as turning a door handle or knob, pressing a doorbell, opening a jar, and using a screwdriver. Following selection of the task, the robot picks up the requisite tool (door knob, doorbell, jar, or screwdriver) and gives it to the patient before simulating the task dynamics and allowing the patient to participate in the task.

The participants of this exploratory study practiced four functional tasks out of six possibilities. These were either selected based on the patient's ability to complete the tasks, or randomly.

Including pre-task selection training, the participants performed 900 sessions with ADAPT in total, with each trial lasting about 1.5 hours and including at least three out of the four tasks, repeated in a random fashion throughout the session. The sessions gradually increased in difficulty over time in accordance with the patients' improving upper body function.

The ADAPT sessions seemed to be effective, as the time needed to complete all the tasks in each session decreased significantly over the study. The robot therapy was also received positively by the patients, with no recorded adverse affects or outside intervention to address robot malfunction being required during the sessions.

Schweighofer and colleagues say that their results "establish the feasibility of ADAPT" and add that the robot system will undergo further testing in a clinical trial.

They conclude in the Journal of NeuroEngineering and Rehabilitation that "safe automatic presentation of functional tasks with ADAPT showed the potential to engage effective motor learning in stroke rehabilitation.

"Motor control and learning principles, such as manipulating the schedule of training, can extend the efficacy of robotic neuro-rehabilitation."

By Helen Albert

Related topics