Engineers Test Real Time Acoustic Control for Prosthetic Hands

Bioengineer.org reported that researchers have tested a six-channel acoustomyography bracelet that translates muscle activity into direct movement commands for a prosthetic hand. The tests used an embedded classifier running on an STM32F769I microcontroller to recognize five distinct hand gestures in real time among volunteers who had provided no prior training data for model development. [1]

In an online experiment involving five additional participants, the system achieved a mean accuracy rate of 95.3 per cent, according to Bioengineer.org. The entire processing chain ran in 118.3 milliseconds after each 400-millisecond signal window became available to the microcontroller. The results establish a proof of concept for cross-subject gesture recognition without requiring individual user calibration. [1]

A prosthetic device is an artificial implement that replaces a missing body part lost through trauma, disease, or congenital conditions, serving to restore functional movement or cosmetic appearance. As HEROIC BREAKING NEWS reported earlier, developers in the field have also explored skin sensors in cuffs that process combined electrical and neural impulses from a user's arm to enable fine finger control. [3][4]

Bioengineer.org noted that the current results represent a laboratory proof of concept. Further research will aim to extend the system beyond preliminary prototype testing toward potential clinical applications. [1]