A wearable vibroarthrography system with PID-controlled negative-pressure suction and synchronized posture estimation.

Abstract

BackgroundVibroarthrography (VAG) is a non-invasive technique for evaluating joint health through acoustic signals generated during movement. However, reliable signal acquisition during dynamic activities remains challenging due to sensor displacement, motion artifacts, and inconsistent skin contact.ObjectiveThis study proposes an advanced wearable VAG system integrating PID-controlled negative-pressure suction and real-time posture estimation to improve signal stability and enhance the assessment of knee joint mechanics during movement.MethodsThe proposed system combines a wearable acoustic sensing module with a vacuum-assisted suction cup and a PID controller that consistently maintains sensor adhesion. Real-time posture estimation is achieved using MediaPipe Pose, which computes knee flexion angles. VAG signals and kinematic data are synchronously acquired during OKC and CKC exercises for integrated analysis.ResultsIn this single-subject pilot test, the system achieved stable sensor attachment and repeatable vibroacoustic recordings throughout dynamic movements. Distinguishable frequency-domain characteristics were observed at relevant knee flexion angles, including approximately 30°, 63°, and 129°.ConclusionsThis pilot feasibility study demonstrates that the proposed wearable VAG platform can achieve stable acquisition of synchronized vibroacoustic and kinematic signals during dynamic knee movement in a single subject. These preliminary findings suggest potential future applications in osteoarthritis screening, rehabilitation assessment, and ambulatory monitoring, pending validation in larger, statistically powered cohorts.