Inferring Fine Finger Motions for Prosthetic Control: An Ultrasound-Based Approach to Real-Time Estimation of Finger...

Area of Science:

• Robotics
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Robotic prostheses currently offer limited functionality, often restricted to discrete gestures.
• Existing systems using electromyography (EMG) or ultrasound (U.S.) struggle with inferring continuous, fine motor movements.
• There is a need for advanced sensor-based systems to improve the dexterity and adoption of prosthetic limbs.

Purpose of the Study:

• To develop an end-to-end robotic system capable of inferring fine, time-evolving finger motions in real-time.
• To bridge the gap between physiological signals and complex motor control in robotic prostheses.
• To enhance the functionality of prosthetic hands for tasks requiring dexterity.

Main Methods:

• Utilized ultrasound (U.S.) imaging to capture muscle dynamics.
• Modeled the human hand as a robotic manipulator for intermediate representation.
• Developed a real-time inference system for fine finger motion detection.

Main Results:

• Successfully demonstrated real-time inference of fine finger motions.
• Validated the system's ability to control prosthetic hand movements for complex tasks.
• Showcased applications in replaying piano music and typing on a keyboard.

Conclusions:

• The developed end-to-end system represents a significant step towards more functional robotic prostheses.
• This technology has the potential to dramatically increase the capabilities and adoption rates of prosthetic limbs for arm amputees.
• This is the first study to demonstrate such fine motor control inference for downstream tasks in an integrated system.