Publication: Data mining approach to assess the significant movement indicators that distinguish people with and without knee oa.
No Thumbnail Available
Date
2023-07-07
Authors
Ko, Jia Liang
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Degeneration of the articular cartilage, the flexible, slick substance that typically shields bones from joint friction and impact, is what constitutes knee osteoarthritis (OA). The disorder can also damage neighboring soft tissues and results in alterations to the bone that lies beneath the cartilage. Limited studies reported on significant movement indicators distinguishing individuals with and without knee OA. Hence, this study aims to investigate the differences in movement patterns specifically knee bending angles between the two categories using data mining approach. Two case studies concerning knee flexion angle & knee radiographic status were employed: from experimental and publicly available data. Numeric data and nominal data were analyzed using statistical analysis (Minitab) and data mining tool (WEKA). Data classification was performed using Node CART classification from Minitab and ZeroR, Naïve Bayes Multinomial Text classifier, LWL, LMT, Random Forest and Decision Stump algorithms to categorize all the attributes into designed class attributes. The main findings reveal that the classification algorithm (Decision Stump) performance accuracy was 93.33% and 86.67% for both Case Study 1 and Case Study 2 respectively. There is a strong relationship of three analysis identified between active knee ROM and knee OA.