The Role of AI in Diagnosing Knee Osteoarthritis with X-Ray Imaging: Today’s Innovations and Tomorrow’s Potential

Machine learning (ML) and deep learning (DL) techniques are playing a growing role in exploring the potential applications of artificial intelligence (AI) in medicine and healthcare. These approaches can be instrumental in providing high-quality care to patients with chronic diseases, such as osteoarthritis. The knee is prone to the most common form of arthritis, which is known as Knee Osteoarthritis (KOA). Degenerative, wear-and-tear arthritis is a common form of arthritis in those over 50 years of age, but it can also develop in younger individuals. Our review of current research gaps and technical challenges in applying AI provides guidance for future investigations in this field. Arthritis has been diagnosed using traditional methods since the 1990s, but with the advancements of AI techniques in the medical field, medical diagnosis has been improved. Henceforth, different ML and DL techniques have been applied by researchers on various medical images like X-ray, Thermal, Ultrasound, and Radiographs, etc. This paper examines the latest and most prominent ML and DL techniques that were published from 2013 to 2023 for detecting KOA severity using X-ray image datasets. The main objective of this study is to present a detailed categorization of major techniques (handcrafted, deep learning based, and hybrid) on the basis of X-ray images used for KOA. Readers will be able to compare and contrast the different techniques, and their accuracy results on KOA severity detection.