Delaying ambulation mode transitions in a powered knee-ankle prosthesis

Area of Science:

• Biomedical Engineering
• Rehabilitation Robotics
• Prosthetics and Orthotics

Background:

• Powered lower limb prostheses offer improved mobility for amputees.
• Seamless mode transitions in multi-modal prostheses are challenging.
• Pattern recognition systems aim for natural transitions but have high error rates.

Purpose of the Study:

• To develop and evaluate a system for powered lower limb prostheses incorporating a delay in mode transitions.
• To assess the impact of a 90 ms transition delay on system reliability and user performance.
• To reduce classification errors in pattern recognition systems for prosthetic mode transitions.

Main Methods:

• Three transfemoral amputees tested a knee-ankle prosthesis across various ambulation modes.
• Mode transitions were evaluated with and without a 90 ms delay following gait events.
• A mode-specific pattern recognition system was trained and tested under both conditions.

Main Results:

• The 90 ms transition delay did not negatively impact user ambulation performance.
• Offline classification errors decreased from 1.30% (non-delayed) to 0.42% (delayed).
• The delayed system demonstrated significantly improved reliability and reduced errors.

Conclusions:

• A small delay in mode transitions can effectively reduce classification errors in powered prostheses.
• This strategy enhances the clinical viability of pattern recognition systems for lower limb prosthetics.
• Delayed transitions represent a promising approach for more reliable prosthetic mobility.