Controlling patient participation during robot-assisted gait training

Area of Science:

• Robotics in Rehabilitation
• Neurorehabilitation Engineering
• Biomechanics of Gait

Background:

• Stroke patients often exhibit passive behavior during robot-assisted gait therapy, limiting rehabilitation benefits.
• Current methods lack effectiveness in optimizing patient activity levels for diverse cognitive and biomechanical impairments.

Purpose of the Study:

• To investigate and develop methods for controlling active patient participation in robot-assisted gait therapy for stroke survivors.
• To address the challenge of passive patient behavior during rehabilitation.

Main Methods:

• Quantified patient activity using heart rate (HR) and weighted sum of interaction torques (WIT) between robot and patient.
• Implemented two control approaches: voluntary effort via visual cues and forced effort by adapting treadmill speed.
• Tested the methods in three experiments with five stroke patients each.

Main Results:

• Successfully controlled patient activity, measured by both WIT and HR, to achieve desired temporal profiles.
• Demonstrated that the control setup can be individually adapted to patients' cognitive and biomechanical needs.

Conclusions:

• Propose a metric to guide clinicians in selecting the optimal patient participation strategy based on individual capabilities.
• Framework facilitates therapists in automatically controlling patient effort to enhance activity levels.
• Increased patient activity is expected to correlate with improved stroke rehabilitation outcomes.