Wireless gyroscope platform enabled by a portable media device for quantifying wobble board therapy

Area of Science:

• Biomedical Engineering
• Rehabilitation Science
• Machine Learning

Background:

• Ankle foot complex rehabilitation often relies on subjective assessments.
• Quantifying therapy progress is crucial for optimizing patient recovery strategies.
• Traditional wobble board therapy lacks objective, remote monitoring capabilities.

Purpose of the Study:

• To develop a system for objective, remote quantification of wobble board therapy for ankle rehabilitation.
• To explore the use of machine learning for classifying ankle status during therapy.
• To enhance patient-therapist interaction and data accessibility in homebound settings.

Main Methods:

• Integration of a wobble board with a portable media device (e.g., iPod) as a wireless gyroscope platform.
• Wireless transmission of gyroscope data for remote analysis.
• Application of a multilayer perceptron neural network for classification of ankle data.
• Consolidation of gyroscope signals into feature sets for machine learning.

Main Results:

• High classification accuracy achieved in differentiating between hemiplegic and unaffected ankles using the developed system.
• Demonstrated feasibility of remote data transmission (e.g., via email attachments).
• Successful integration of wireless technology with conventional therapy equipment.

Conclusions:

• The combined approach of machine learning, wireless systems, and wobble boards significantly advances ankle rehabilitation capabilities.
• This technology enables objective, data-driven therapy adjustments and improved remote patient care.
• The system offers a promising solution for optimizing rehabilitation outcomes in homebound patients.