Design and validation of a novel 3D-printed wearable device for monitoring knee joint kinematics. (August 2021)

• Record Type: Journal Article
• Authors: Young, Calvin; Oliver, Michele L.; Gordon, Karen D.
• Is Part Of: Medical engineering & physics. Volume 94(2021)
• Page Start: 1
• Page End: 7
• Publication Date: 2021-08
• Subjects: Knee kinematics -- Gait -- Wearable sensor -- Validation; Biomedical engineering -- Periodicals; Biomedical Engineering -- Periodicals; Physics -- Periodicals; Génie biomédical -- Périodiques; Biomedical engineering; Electronic journals; Periodicals; 610.28
• DOI: 10.1016/j.medengphy.2021.05.013
• Languages: English
• ISSNs: 1350-4533

Abstract

Highlights: A novel 3d printed wearable device was developed for monitoring knee kinematics. The wearable device measures knee flexion/extension and adduction/abduction angles. Reported angles were validated against a gold standard during treadmill walking. The wearable showed low error, high correlation, and good repeatability. Performance was comparable with other custom and commercially available wearables. Abstract: Gait analysis provides an important tool for the study and clinical evaluation of conditions which affect knee joint biomechanics. Collection of knee joint kinematics in real world environments during locomotor activities of daily living could provide quantitative evidence to help understand functional impairment. Unfortunately, the high cost and necessary technical expertise associated with current commercially available systems for kinematic monitoring serve as an impediment to their adoption outside of specialized research groups. We have developed a low-cost, custom wearable device to address these shortcomings. The 3D printed device is capable of measuring knee flexion/extension (F/E) and adduction/abduction (AD/AB) angles. Here, we present a gold standard validation of the novel device against an optoelectronic motion capture system (MCS). Data were collected during a treadmill walking task from 8 participants on 2 separate occasions. Agreement with the MCS was quantified via root mean squared error (RMSE), coefficients of multiple correlation (CMC), paired dependent t -tests and Bland-Altman analyses. The wearable device had an overall RMSE of 3.0 ° and 2.7 ° and a CMC of 0.97 and 0.91 in F/E and AD/AB respectively. Wearable device error showed no significant differences between test occasions, and Bland-Altman analyses showed low bias with narrow limits of agreement. These results demonstrate the capability of the device to accurately and reliably monitor knee F/E and AD/AB angles showing strong potential for field implementation.