Search for a command to run...
The risk of developing symptomatic knee osteoarthritis (KOA) during a lifetime, i.e., pain, aching, or stiffness in a joint associated with radiographic KOA, was estimated in 2008 to be around 40% in men and 47% in women. The clinical and scientific communities lack an efficient diagnostic method to effectively monitor, evaluate, and predict the evolution of KOA before and during the therapeutic protocol. In this review, we summarize the main methods that are used or seem promising for the diagnosis of osteoarthritis, with a specific focus on non- or low-invasive methods. As standard diagnostic tools, arthroscopy, magnetic resonance imaging (MRI), and X-ray radiography provide spatial and direct visualization of the joint. However, discrepancies between findings and patient feelings often occur, indicating a lack of correlation between current imaging methods and clinical symptoms. Alternative strategies are in development, including the analysis of biochemical markers or acoustic emission recordings. These methods have undergone deep development and propose, with non- or minimally invasive procedures, to obtain data on tissue condition. However, they present some drawbacks, such as possible interference or the lack of direct visualization of the tissue. Other original methods show strong potential in the field of KOA monitoring, such as electrical bioimpedance or near-infrared spectrometry. These methods could permit us to obtain cheap, portable, and non-invasive data on joint tissue health, while they still need strong implementation to be validated. Also, the use of Artificial Intelligence (AI) in the diagnosis seems essential to effectively develop and validate predictive models for KOA evolution, provided that a large and robust database is available. This would offer a powerful tool for researchers and clinicians to improve therapeutic strategies while permitting an anticipated adaptation of the clinical protocols, moving toward reliable and personalized medicine.