Abstract
Clinicians lack objective means for monitoring if their knee osteoarthritis patients are improving outside of the clinic (e.g., at home). Previous human activity recognition (HAR) models using wearable sensor data have only used data from healthy people and such models are typically imprecise for people who have medical conditions affecting movement. HAR models designed for people with knee osteoarthritis have classified rehabilitation exercises but not the clinically relevant activities of transitioning from a chair, negotiating stairs and walking, which are commonly monitored for improvement during therapy for this condition. Therefore, it is unknown if a HAR model trained on data from people who have knee osteoarthritis can be accurate in classifying these three clinically relevant activities. Therefore, we collected inertial measurement unit (IMU) data from 18 participants with knee osteoarthritis and trained convolutional neural network models to identify chair, stairs and walking activities, and phases. The model accuracy was 85% at the first level of classification (activity), 89–97% at the second (direction of movement) and 60–67% at the third level (phase). This study is the first proof-of-concept that an accurate HAR system can be developed using IMU data from people with knee osteoarthritis to classify activities and phases of activities.
Highlights
Osteoarthritis is one of the leading causes of disability [1,2]
We have demonstrated a proof-of-concept that data collected from people who have a confirmed diagnosis of knee osteoarthritis can feasibly be used to train a human activity recognition (HAR) model using convolutional neural networks (CNN)
Previous literature reporting the development of HAR models for people with knee osteoarthritis had not explored (1) the potential accuracy of a model trained on data collected from people who have knee osteoarthritis rather than healthy people, and (2)
Summary
Osteoarthritis is one of the leading causes of disability [1,2]. People with knee osteoarthritis have symptoms such as pain and stiffness that result in difficulty performing specific physical activities such as transitioning from a chair, negotiating stairs [3] and walking [4]. While there is growing interest in wearable sensor technology for use in clinical environments, no studies have investigated if machine learning approaches can assist with monitoring improvement in the performance of clinically relevant activities outside of a clinical environment, for patients diagnosed with knee osteoarthritis [5]. There is high-quality evidence of improvements in pain and function following movement interventions, such as exercise, or surgical interventions. For people receiving these treatments, the leading medical society dedicated to researching osteoarthritis, the Osteoarthritis Research Society International, recommends that people who have a confirmed diagnosis of knee osteoarthritis are monitored for improvement in the performance of three specific and clinically relevant everyday activities; (1) transitioning from a chair, 4.0/)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
