Quantifying Rheumatoid Arthritis Disease Activity Using a Multimodal Sensing Knee Brace

Area of Science:

• Biomedical Engineering
• Rheumatology
• Wearable Technology

Background:

• Rheumatoid arthritis (RA) is a chronic inflammatory joint disease.
• Current RA monitoring relies on subjective patient and physician assessments.
• Objective, quantitative methods are needed to track RA disease activity and treatment efficacy.

Purpose of the Study:

• To develop and validate a noninvasive sensing system for quantitative monitoring of RA disease activity.
• To assess the feasibility of using joint acoustic emissions (JAEs), electrical bioimpedance (EBI), and kinematics for RA monitoring.
• To provide objective data for improved RA management.

Main Methods:

• A custom multimodal sensing brace was used to collect JAEs, EBI, and kinematic data from 11 RA patients over three weeks.
• Features from the sensing modalities were analyzed using linear discriminant analysis (LDA) and ridge regression.
• The system predicted disease activity scores (DAS28-ESR) and erythrocyte sedimentation rate (ESR) levels.

Main Results:

• The system achieved an area under the receiver operating characteristic curve (AUC) of 0.82 for predicting DAS28-ESR.
• JAEs alone effectively tracked intrasubject disease activity differences and classified ESR levels with an AUC of 0.93.
• Patients expressed interest and ability in using the brace for at-home monitoring.

Conclusions:

• Noninvasive sensing, particularly JAEs, can accurately detect and monitor RA disease activity.
• The developed system offers a potential for objective, quantitative RA disease activity assessment.
• This technology could enhance RA management by providing data independent of clinical visits.