A Comprehensive Review of Machine Learning’s Role within KOA

Abstract

INTRODUCTION: Knee Osteoarthritis (KOA) is a degenerative joint disease, that predominantly affects the knee joint and causes significant global disability. The traditional methods prevailing in this field for proper diagnosis are very subjective and time-consuming, which hinders early detection. This study explored the integration of artificial intelligence (AI) in orthopedics, specifically the field of machine learning (ML) applications in KOA. OBJECTIVES: The objective is to assess the effectiveness of Machine learning in KOA, besides focusing on disease progression, joint detection, segmentation, and its classification. ML algorithms are also applied to analyze the MRI and X-ray images for their proper classification and forecasting. The survey spanning from 2018 to 2022 investigated the treatment-seeking behavior of individuals with OA symptoms. METHODS: Utilizing deep learning (CNN, RNN) and various ML algorithms (SVM, GBM), this study examined KOA. Machine learning was used as a subset of AI, and it played a pivotal role in healthcare, particularly in the field of medical imaging.  The analysis involved reviewing the studies from credible sources like Elsevier and Web of Science. RESULTS: Current research in the field of medical imaging CAD revealed promising outcomes. Studies that utilized CNN demonstrated 80-90% accuracy on datasets like OAI and MOST, emphasizing its varied significance in vast clinical and imaging data archives. CONCLUSION: This comprehensive analysis highlighted the evolving landscape of research in KOA. The role of machine learning in classification, segmentation, and diagnosis of severity is very much evident. The study also anticipates a future framework optimizing KOA detection and overall classification performance, with a strong emphasis on the potential for enhancement of knee osteoarthritis diagnostics.