Novel evaluation of upper-limb motor performance after stroke based on normal reaching movement model

Area of Science:

• Robotics in rehabilitation medicine
• Neurorehabilitation and motor control
• Biomechanical analysis of human movement

Background:

• Upper-limb rehabilitation robots assist post-stroke patients with repetitive reaching movements.
• Optimizing robot-aided training requires personalization based on individual motor characteristics.
• Current methods lack objective evaluation against an individual's pre-stroke motor performance.

Purpose of the Study:

• To develop and validate a novel method for evaluating upper limb motor performance in stroke survivors.
• To establish an objective evaluation standard based on an individual's normal reaching movement model.
• To enable individualized robot-aided training protocols by assessing unique motor characteristics.

Main Methods:

• Candidate models including Fitts' law and the Almanji model were considered for representing normal reaching performance.
• Kinematic data from healthy (n=12) and post-stroke (n=7) subjects were collected using a robot.
• A pilot study (n=12) used the less-affected arm's performance to model and predict normal reaching for evaluating the affected arm.

Main Results:

• The proposed normal reaching model accurately identified reaching patterns in healthy (n=12) and less-affected arms (n=19, R² > 0.7).
• The model effectively distinguished between normal and erroneous reaching movements of the affected arm.
• The evaluation method visually and intuitively highlighted individual motor characteristics of affected upper limbs.

Conclusions:

• The developed method provides an individualized assessment of reaching characteristics using a personal normal reaching model.
• This approach facilitates personalized rehabilitation by identifying specific reaching movements for targeted training.
• The method holds potential for optimizing robot-aided upper limb rehabilitation for stroke patients.