Knee Osteoarthritis Severity Classification using SVM

Numerous ailments have adversely affected people's lifestyles in recent years. Among these, osteoarthritis (OA) of the knee has been identified as the main source of disability and restriction of activity, especially in senior citizens. It is essential to create fast and cost-efficient methods for diagnosing knee OA. In this paper, we analyze severity of OA in knee X-ray images by employing two machine learning (ML) models: decision tree (DT), as well as support vector machine (SVM). For capturing characteristics of the knee, we utilized histogram of oriented gradients (HOG) technique. Classifiers benefit from an enriched feature space. According to empirical findings, the SVM classifier achieved a 70% accuracy rate, outperforming the other.