Control interfaces for intention detection in active transfemoral prosthetics: a systematic review

Area of Science:

• Biomedical Engineering
• Rehabilitation Engineering
• Prosthetics and Orthotics

Background:

• Active transfemoral prosthetic devices require sophisticated control interfaces for effective user intention detection.
• Existing literature spans two decades, focusing on control algorithms for knee prostheses.

Purpose of the Study:

• To review and classify the latest methods for creating control interfaces for intention detection in active transfemoral prosthetic devices.
• To assess the feasibility of current control interfaces and identify key research areas.

Main Methods:

• A comprehensive literature review was conducted over the past two decades.
• 213 studies were initially retrieved, with 33 selected based on relevance and quality using the Downs and Black checklist.
• Studies were categorized into control strategy frameworks, conventional methods, implementation potentials, and machine learning algorithms.

Main Results:

• Fifteen papers focused on control strategy frameworks, two on conventional methods, four on implementation potentials, and twelve on machine learning algorithms.
• Machine learning algorithms were a significant area of investigation for active prosthetic legs.
• The review identified a need for simpler sensory systems paired with innovative control algorithms.

Conclusions:

• Effective sensory systems and intention detection algorithms are critical for advanced transfemoral prosthetic limbs.
• Integrating simpler sensory systems with advanced algorithms can enhance the translation of sensor data.
• The development of feasible control interfaces for intention detection is crucial for improving active prosthetic leg functionality.