Osteoarthritis is the most common joint disease in the world. It can occur in any joint, but it is the most common in hand, knee, hip and spine. Osteoarthritis is a whole joint disease affecting simultaneously several joint tissues, i.e. articular cartilage, subchondral bone, meniscus, synovium, ligaments and tendons. The typical primary signs of osteoarthritis progression are degeneration and wear of articular cartilage along with pathological remodeling of the subchondral bone.

During the last decades, we have seen the rapid development of different imaging modalities and digital image analysis methods both at the laboratory level, i.e. tissue and cell level, and at the clinical level. This development has allowed both researchers and clinicians to better understand the initiation and progression of osteoarthritis. Specifically, machine learning based approaches for image analysis have become more common and promising during the recent few years.

In this talk, the role of several imaging modalities in osteoarthritis research and clinical diagnostics - along with advanced image analysis methods - will be introduced. From the laboratory imaging methods, we will focus micro-computed tomography (micro-CT), Fourier-transform infrared imaging (FTIRI), Raman microscopic imaging, and polarized light microscopy (PLM). From the clinical imaging methods, we will focus on conventional radiography (X-ray) and the potential of advanced image analysis and deep learning algorithms to mine new diagnostic and prognostic information from them. Finally, the future prospects of clinical prediction models, combining imaging data and clinical information, will be discussed.