Design and Evaluation of a Wrist Wearable Joint Acoustic Emission Monitoring System

Abstract

Joint acoustic emission (JAE) sensing is emerging as a potential modality for quantitative at-home joint health assessment. We designed and validated a low-profile, easy-to-use wearable system for JAE sensing at the wrist. An embedded microcontroller on a wrist-worn printed circuit board is used to record multi-microphone (mic) joint acoustics sampled at 46.875 kHz using an on-board analog-to-digital converter. A flex sensor and a force sensitive resistor (FSR) are sampled at 300 Hz to capture kinematics and mic backing force. Custom sensor casing solutions and real-time user feedback systems enhance audio sensing capabilities at locations both proximal and distal to the wrist. An experiment extracting wrist JAEs from healthy adults (n=6) allowed for comparison to a previously established benchtop JAE sensing system. The acoustic data were bandpass filtered (150 Hz-5.5 kHz). Qualitative observations reveal the wearable system mics successfully capture JAEs. Signal-to-noise ratio (SNR), intraclass correlation coefficient (model 3, k), and coefficients of variability were calculated to evaluate acoustic signal strength and repeatability in both systems' recordings. SNR reveals that JAE acoustic signal strength is higher when recorded using the wearable system than with the benchtop system (p<0.01). Reliability measures show that the wearable system records JAEs with similar levels of reliability to the benchtop system. Initial recordings in clinical wrist JAE research studies demonstrate high quality JAE measurements from children with Juvenile Idiopathic Arthritis (JIA) and healthy controls (HCs) (4 JIA, 3 HC). Eventually, this system may be developed into a tool for at-home wrist joint health monitoring.