Background/Purpose:

Juvenile Idiopathic Arthritis (JIA) is the most common, chronic, childhood rheumatic disease and an important cause of disability. It is characterized by persistent inflammation of the joints, with onset in childhood. A pronounced improvement in functional outcome has been documented in the past decade – largely due to advances in clinical diagnosis and treatment regimes. However, 5-10% of patients still have serious functional disability five years post diagnosis. Outcomes will be further improved by advances in the standardization of clinical diagnostics, disease tracking, and further refining of the treatment regimen.

The diverse nature of JIA, the lack of specific biomarkers and its concomitant progression have made standardizing diagnosis and tracking of the disease difficult. Currently, diagnosing and tracking JIA relies on subjective physical exams, reported symptoms, and imaging studies. There exists a need for a non-invasive, safe, radiation-free technique for objectively and accurately assessing the status of affected joints.

Methods:

Joint sounds can be used to quantitatively assess the joint. Microphones can be used to capture information regarding the underlying physiologic processes of a joint. This technique is safe, radiation-free, objective, and quantitative. The joint sounds change with the biomechanics of the joint: internal friction between articulating structures in the knee create various frequencies of vibrations that can be detected at the surface of the knee. The inflammation and altered biomechanical properties provide the opportunity for utilizing this technology to diagnose and monitor the condition.

We performed joint sound recordings on four JIA patients and seven healthy controls that were age and sex matched. The knee sounds of the subjects were recorded using a custom hardware setup consisting of LED motion tracking and accelerometers for vibration detection (Figure 1A shows a leg in flexion with recording apparatus). The subjects performed ten flexion/extension cycles with the recording apparatus (1B exercise and microphone schematic).

Results:

 The results can be seen in Figure 1C. The top graph shows the angle of the knee, while the bottom plot shows the sound profile. The plots of JIA patients are more chaotic with periodic peaks. After treatment, the sounds of JIA patient more closely resembled the knee sounds of quiet healthy control.  Signal analysis was performed between two groups. We found that RMS power and the standard deviation of the signals were substantially different between the groups (1D).

Conclusion:

The substantial differences in joint sound signals between healthy children and children with JIA are encouraging for the further development and refinement of this acoustic recording system and associated knee health algorithm for the diagnosis and objective monitoring of JIA.

Figure 1 Experimental Setup (A,B), Time domain signal (C) and Signal Comparison (D).