A Proof-of-Concept System to Analyze Joint Sounds in Real Time for Knee Health Assessment in Uncontrolled Settings

A proof-of-concept wearable system for measuring, processing, analyzing, and logging activity-contextualized joint sound signatures from the knee joint is presented. Microelectro-mechanical systems (MEMS)-based microphones are used to detect the acoustical emissions from the knee joint, and MEMS accelerometer-gyroscope pairs at the joint are used to calculate joint angle. The joint angle measurement is used as a context for evaluating the resultant acoustical emissions of the knee joint during unloaded flexion-extension cycles. Automated click detection, performed real-time on-board the field-programmable gate array, is demonstrated successfully in both quiet (lab) and simulated loud (coffee shop) environments for proof-of-concept recordings.