Sparse Wearable Sonomyography Sensor-based Proprioceptive Proportional Control Across Multiple Gestures

TL;DR

The wearable SMG system provided accurate, stable, and intuitive control in real time by achieving an average success rate greater than 80% with all gestures and provided insights to validate SMG as an intuitive human–machine interface.

Abstract

Sonomyography (SMG) is a non-invasive technique that uses ultrasound imaging to detect the dynamic activity of muscles. Wearable SMG systems have recently gained popularity due to their potential as human-computer interfaces for their superior performance compared to conventional methods. This paper demonstrates real-time positional proportional control of multiple gestures using a multiplexed 8-channel wearable SMG system. The amplitude-mode ultrasound signals from the SMG system were utilized to detect muscle activity from the forearm of 8 healthy individuals. The derived signals were used to control the on-screen movement of the cursor. A target achievement task was performed to analyze the performance of our SMG-based human-machine interface. Our wearable SMG system provided accurate, stable, and intuitive control in real-time by achieving an average success rate greater than 80% with all gestures. Furthermore, the wearable SMG system's abilities to detect volitional movement and decode movement kinematic information from SMG trajectories using standard performance metrics were evaluated. Our results provide insights to validate SMG as an intuitive human-machine interface.

Figures (12)

• Figure 1: a) Piezoceramic transducer element with an optimized backing layer, b) Exploded rendering of the packaged transducer showing the matching network and 3D printed housing, c) 8-channel sensor array on a wearable velcro band, d) Time and frequency domain response from pulse-echo tests of the sensor.
• Figure 2: a) Experimental setup: the participants were instrumented with a custom 8-channel SMG sensor connected to a commercial ultrasound pulser-receiver system. The participants performed a real-time target achievement task. b) A-mode signal preprocessing: The preprocessed A-mode signals from all 8 channels were arranged row-wise to form a sparse ultrasound frame. c) Generation of cursor control signal: the control signal was generated by quantifying the gesture position using Pearson's correlation coefficient.
• Figure 3: Example of user trajectory and presented target. The outcome metrics, namely, movement time ($T_m$), endpoint error, endpoint stability, path efficiency, and maximum velocity, were calculated from user trajectory.
• Figure 4: Relationship between MCP joint angle and SMG signal for three gestures with the sensor array placed at proximal and distal locations on the forearm. Plots show the best $R^2$ values obtained for each gesture at each sensor position.
• Figure 5: The movement trajectories of a representative participant for a target presented at 0.6 achieved using all gestures