Abstract:Imaging plays a pivotal role in osteoarthritis research, particularly in epidemiological and clinical trials of knee osteoarthritis (KOA), with the ultimate goal being the development of an effective drug treatment for future prevention or cessation of disease. Imaging assessment methods can be semi-quantitative, quantitative, or a combination, with quantitative methods usually relying on software to assist. The software generally attempts image segmentation (outlining of relevant structures). New techniques using artificial intelligence (AI) or deep learning (DL) are currently a frequent topic of research. This review article provides an overview of the literature to date, focusing primarily on the current status of quantitative software-based assessment techniques of KOA using magnetic resonance (MR) imaging. We will concentrate on the imaging evaluation of three specific structural imaging biomarkers: bone marrow lesions (BMLs), meniscus, and synovitis consisting of effusion synovitis (ES) and Hoffa’s synovitis (HS). A brief clinical and imaging background review of osteoarthritis evaluation, particularly relating to these three structural markers, is provided as well as a general summary of the software methods. A summary of the literature with respect to each KOA assessment method will be presented overall as well as with respect to each specific biomarker individually. Novel techniques, as well as future goals and directions using quantitative imaging assessment, will be discussed.

Quantitative bone marrow lesion, meniscus, and synovitis measurement: current status