A Support Vector Regression Approach for Continuous Prediction of Ankle Angle and Moment During Walking: An...

Area of Science:

• Biomedical Engineering
• Rehabilitation Robotics
• Biomechanics

Background:

• Lower limb amputation significantly impacts locomotion, necessitating advanced prosthetic solutions.
• Current passive prostheses have limited functionality; active prostheses offer enhanced motor support.
• Effective control of active prostheses requires accurate interpretation of user's locomotive intentions.

Purpose of the Study:

• To develop a continuous estimation strategy for ankle angle and ankle moment in transtibial amputees.
• To enhance the control of active ankle prostheses for restoring natural gait.
• To improve prosthetic limb functionality by predicting user's intended movement.

Main Methods:

• Utilized a support vector regression (SVR) approach for continuous estimation.
• Employed experimentally obtained hip and knee joint motion data as input features.
• Validated the SVR model for predicting ankle kinematics and kinetics during walking.

Main Results:

• The proposed SVR method accurately predicted ankle angle (mean R² = 0.98).
• High accuracy was also achieved in predicting ankle moment (mean R² = 0.97).
• The model demonstrated effectiveness for level ground walking at self-selected speeds.

Conclusions:

• The developed SVR strategy provides a reliable method for estimating critical ankle parameters.
• This approach holds significant potential for improving the control algorithms of active transtibial prostheses.
• Accurate estimation can lead to more intuitive and natural locomotion for amputees.