Deep Learning and IoT-Based Ankle-Foot Orthosis for Enhanced Gait Optimization

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Artificial Intelligence in Healthcare

Background:

• Gait imbalances affect mobility and quality of life.
• Ankle-foot orthosis (AFO) devices provide crucial support for individuals with muscle weakness or paralysis.
• Current AFO management lacks personalized, data-driven insights.

Purpose of the Study:

• To develop an integrated Internet of Things (IoT) and machine learning system for managing gait imbalances.
• To revolutionize medical orthotics with a sophisticated, data-driven approach.
• To enhance patient care through personalized gait management solutions.

Main Methods:

• A smart ankle-foot orthosis (AFO) integrated with surface electromyography (sEMG) and Inertial Measurement Unit (IMU) sensors.
• Cloud-based data transmission via fog computing for real-time analysis.
• Machine learning models including Transformer, LSTM, ANN, SVM, Random Forest, and Decision Tree were employed for gait phase classification.

Main Results:

• The Transformer model achieved 98.97% accuracy in classifying normal versus aberrant walking phases.
• Accurate prediction and measurement of patient walking pattern improvements were demonstrated.
• The system effectively distinguished between normal and aberrant gait phases.

Conclusions:

• Predictive analytics enable tailored recommendations for AFO usage duration and intensity.
• Physician validation and patient access to progress reports ensure continuous monitoring and treatment adjustments.
• This approach facilitates personalized rehabilitation and improved patient outcomes.