Adaptive Robotic System for Stroke Rehabilitation
Author Information
Author(s): Kan Patricia, Huq Rajibul, Hoey Jesse, Goetschalckx Robby, Mihailidis Alex
Primary Institution: University of Toronto
Hypothesis
Can a robotic system using a decision theoretic model improve upper-limb rehabilitation for stroke patients?
Conclusion
The study suggests that the robotic system's decisions were generally aligned with those of a human therapist, indicating potential for clinical use.
Supporting Evidence
- The therapist agreed with the system's decisions approximately 65% of the time.
- The patient expressed satisfaction with the robotic system and would use it as their primary rehabilitation method.
- The system's decisions were generally viewed as believable by the therapist.
Takeaway
This study tested a robot that helps stroke patients practice reaching exercises, and it mostly agreed with what a therapist would decide.
Methodology
The system's performance was evaluated by comparing its decisions with those of a human therapist over six sessions with one patient.
Potential Biases
The system may have overestimated user fatigue, leading to premature stopping of exercises.
Limitations
The sample size was limited to one patient and one therapist, restricting the generalizability of the findings.
Participant Demographics
One right-side hemiparetic stroke patient, 227 days post-stroke.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website