Osteoarthritis severity using linear vector quantization based first order feature

Abstract

Many researchers conducted research on osteoarthritis. This is due to the large number of osteoarthritis patients. In Indonesia, one of ten people at risk of osteoarthritis. In addition, osteoarthritis cannot be cured, so it is important to know its status earlier. This study only focuses on the Decision Support System on the knee although osteoarthritis can occur in the hip, spine, thumb, index finger and toe. The severity of osteoarthritis which is divided into 5 clusters, namely KL-Grade 0 to KL-Grade 4. KL-Grade 0 shows normal conditions, and KL-Grade 4 is the worst condition. The purpose of this study is to build a Decision Support System (DSS) to determine osteoarthritis severity using Linear Vector Quantization (LVQ) based on First Order (FO) features. The method used is Linear Vector Quantization (LVQ) based on First Order (FO) features. The experiment was divided into four stages: image processing, feature extraction learning process, and testing process. The results obtained were that the system can classified well for KL-Grade 4 and KL-Grade 3, while for KL-Grade 0, KL-Grade 1, and KL-Grade 3 it still cannot properly qualify according to the cluster.