Comparison of Wearable and Computer Vision Based Approaches to Knee Flexion Angle Measurement

This paper outlines the development of a wearable device to measure knee joint flexion angles and compares its performance to a computer vision based human pose model. The objective is to advance techniques used in measuring joint flexion angles, which have several drawbacks such as relying heavily on the clinician for accuracy while also being invasive as they require physical contact between the patient and orthopaedic specialist. The proposed implementations integrate digital innovation in orthopaedic healthcare while being affordable and easy to deploy, making them marketable to clinicians and patients in developing countries. To evaluate the workability of the implementations, knee flexion angle measurements of a test subject were obtained using both techniques which yielded consistent and comparable flexion angle readings. Further development of this work can have a significant impact on the accessibility and affordability of innovative healthcare monitoring technologies, especially for patients in developing countries.