Automated Quantification of Movement Qualities in the Human Upper Extremity After Stroke Using a Wearable Robot.

Area of Science:

• Biomedical Engineering
• Neurorehabilitation
• Robotics

Background:

• Stroke is a primary cause of long-term adult disability, frequently resulting in upper extremity (UE) motor impairments.
• Conventional assessments like the Fugl-Meyer Assessment (FMA) use subjective, ordinal scales, limiting objective motor quality evaluation.
• Wearable robotics offer high-resolution data but lack standardized, clinically interpretable protocols for stroke rehabilitation.

Purpose of the Study:

• To develop an objective, standardized, and clinically interpretable method for quantifying UE motor qualities.
• To integrate wearable robotic technology with traditional clinical assessment tasks for improved stroke motor assessment.

Main Methods:

• Ten healthy individuals and ten stroke survivors performed standardized UE tasks wearing the HARMONY exoskeleton.
• A "trajectory pattern similarity score" was developed using root mean square error against normative joint trajectories.
• Kinematic synergy analysis via non-negative matrix factorization evaluated multi-joint coordination alterations.

Main Results:

• The trajectory pattern similarity score strongly correlated with clinical FMA-UE scores (r = -0.93) and showed high test-retest reliability (ICC = 0.98).
• A significant decrease in identified kinematic synergies correlated with increased motor impairment severity (r = 0.79).
• Kinematic synergy analysis revealed mechanisms of impaired motor control, including pathological joint coupling and loss of individual joint control.

Conclusions:

• A novel, standardized assessment framework integrating wearable robotics with clinical tasks was presented.
• This approach bridges objective robotic data and clinical interpretability for robust motor impairment assessment.
• The framework enables intuitive phenotyping of motor characteristics to guide personalized stroke rehabilitation strategies.