Journal of IMAB (Oct 2024)
THE ROLE POTENTIAL OF ARTIFICIAL INTELLIGENCE IN KNEE OSTEOARTHRITIS
Abstract
Purpose: The purpose of the study is to examine available scientific sources related to the role and potential of artificial intelligence in osteoarthritis of the knee joint. Materials/Methods: Method of deduction (analysis of literary sources). To achieve the goal, available scientific data on the role and potential of artificial intelligence application in knee OA were studied and analyzed. Results: The following innovations related to the use of artificial intelligence in knee osteoarthritis (OA) were reviewed: artificial intelligence (AI) software - named KOALA™ and DL AI software - MediAI-OA. KOALA™ is software that provides metric evaluations of knee joint imaging. Standardized quantitative measurements of morphological features such as joint gap width and joint gap area on knee radiographs reduce errors in diagnosis. The new DL software, MediAI-OA, demonstrated good success rates in analyzing knee OA characteristics, Kellgren-Lawrence (KL) grading (which is used to classify the severity of knee OA), and OA diagnosis comparable to that of experienced orthopedists and radiologists. Discussion: Diagnostic imaging is a vital tool for visualization. Imaging methods such as radiography, magnetic resonance (MR), computed tomography (CT), and ultrasound play critical roles in OA diagnosis. Additionally, vibro- and phono arthrography serve as alternative diagnostic tools. The most commonly used imaging method is magnetic resonance imaging, which has been found to underestimate the extent of osteochondral lesions. This can lead to inadequate and incomplete diagnoses. Artificial intelligence can serve as a strategic element in addressing these limitations in radiographic knee OA diagnosis. Conclusion: Artificial intelligence has the potential to advance the field of radiology by enhancing efficiency, accuracy, and precision in the radiographic diagnosis of knee osteoarthritis.
Keywords
