Unveiling Risk Factors of Knee Osteoarthritis: A Novel Approach of Integrating Sample Survey Insights using Statistical and Machine Learning Model

This survey is aimed to identify risk factors associated with knee osteoarthritis (KOA) among various occupational categories in Pune and surrounding regions. The survey was conducted using multiple media platforms, including email, face-to-face interviews, and telephonic conversations, to ensure broad participation. Statistical analysis based on Confidence Intervals, Chi-square tests, and Spearman's correlation coefficient revealed significant risk factors of knee OA among various parameters such as gender, age, BMI, followed by subcategories of occupational activities, knee movement activities, and functionality. In addition, machine learning algorithms, such as Support Vector Machine (SVM) and Random Forest (RF) classifiers, were employed to find the probable association between knee pathological parameters and knee pain. The SVM (RBF) model demonstrated higher prediction accuracy of 88% compared to the RF model. One notable finding of this study is that individuals engaged in regular sports activities showed a reduced risk of knee pain. The identification of the impact of age, BMI on OA development, the influence of prolonged standing, and heavy physical work on knee pain have significant implications for preventive measures underscore the importance of early detection, appropriate workplace modifications, and promoting regular physical activity to lower the burden of knee OA.