Estimating Quality of Reaching Movement Using a Wrist-Worn Inertial Sensor

Area of Science:

• Neuroscience
• Rehabilitation Engineering
• Biomedical Signal Processing

Background:

• Stroke significantly impacts motor function, leading to long-term disability.
• Current remote monitoring often quantifies movement amount, not quality, limiting therapy assessment.
• Accurate, unobtrusive measurement of motor performance quality is crucial for effective stroke rehabilitation.

Purpose of the Study:

• To develop and validate a novel method for measuring the quality of individual reaching movements in stroke survivors.
• To leverage motor control theory for seamless and unobtrusive assessment of motor performance.
• To estimate the Functional Ability Scale (FAS) scores of reaching movements remotely.

Main Methods:

• Utilized a motor control theory framework to analyze reaching movements.
• Developed a system to measure the quality of individual reaching movements unobtrusively.
• Trained and evaluated the system on 19 stroke survivors to estimate FAS scores.

Main Results:

• The system successfully estimated Functional Ability Scale (FAS) scores for reaching movements, with minor confusion between adjacent scores.
• Overall Normalized Root Mean Square Error (NRMSE) for subject average FAS scores was 22.5%.
• Excluding severe cases, the NRMSE for mild-to-moderate subjects was 13.1%, indicating high accuracy.

Conclusions:

• The proposed method effectively estimates the quality of reaching movements and functional ability in stroke survivors.
• This technique holds promise for remote patient monitoring in home and community settings.
• Further development could enhance accuracy, particularly for individuals with severe motor impairments.