Quantitative Evaluation of Performance during Robot-assisted Treatment.

Area of Science:

• Rehabilitation Engineering
• Robotics in Medicine
• Pediatric Physical Therapy

Background:

• Robotic systems offer significant potential for collecting quantitative data in clinical settings.
• This data is often underutilized in patient rehabilitation due to a lack of simple, integrated assessment tools.
• Developing automated metrics is crucial for efficient and objective patient monitoring.

Purpose of the Study:

• To introduce a straightforward parameter (P) for assessing subject performance during upper limb robotic training.
• To leverage robot-recorded data for performance evaluation without additional patient or clinician burden.
• To provide a quantitative measure of progress in robotic-assisted rehabilitation.

Main Methods:

• Fourteen children with cerebral palsy (CP) participated in upper limb training using the Armeo®Spring robotic device.
• Performance during each session was evaluated using the parameter P, derived from automatically recorded robot data.
• Group analysis involved computing median and interquartile values of P over time.

Main Results:

• The study observed a significant increase in the median (interquartile) values of parameter P from 0.27 (0.21) at baseline (T0) to 0.55 (0.27) at follow-up (T1).
• This quantitative improvement in robotic training performance was corroborated by a significant enhancement in the Melbourne Assessment of Unilateral Upper Limb Function.
• The results demonstrate the parameter's sensitivity to changes in motor function.

Conclusions:

• The developed parameter P effectively quantifies performance variations during robotic training.
• It provides a reliable and objective measure of motion abilities, comparable to established clinical scales.
• This simple, automated parameter enhances the clinical utility of robotic rehabilitation data.