Physical activity as a risk factor in the progression of osteoarthritis: a machine learning perspective

Abstract

Knee osteoarthritis (KOA) comes with a variety of symptoms’ intensity, frequency and pattern. Most of the current methods in KOA diagnosis are very expensive commonly measuring changes in joint morphology and function. So, it is very important to diagnose KOA early, which can be achieved with early identification of significant risk factors in clinical data. Our objective in this paper is to investigate the predictive capacity of physical activity measures as risk factors in the progression of KOA. In order to achieve this, a machine learning approach is proposed here for KOA prediction using features extracted from an accelerometer bracelet. Various ML models were explored for their suitability in implementing the learning task on different combinations of feature subsets. Results up to 74.5% were achieved indicating that physical activity measured by accelerometers may constitute an important risk factor for KOA progression prediction especially if it is combined with complementary data sources.