Purpose: Osteoarthritis (OA) is the most common degenerative joint disease and the leading cause of pain and disability worldwide. OA is characterised by a loss of articular cartilage, subchondral bone thickening and osteophytosis, as well as changes to the synovium, meniscus, tendons/ligaments and muscles. OA is a disease showing slow progression and is thought to develop over a period of 10-20 years. Changes at the cellular level develop over time with symptomatic OA becoming evident only with established disease in many cases. Early diagnosis is therefore challenging, but essential to ensure delivery of treatments in a timely manner to modify the course of the disease and produce better patient outcomes. Current diagnostic tools for OA include radiography, magnetic resonance imaging (MRI) and arthroscopic optical coherence tomography (OCT). OCT in particular, can produce high-resolution, cross sectional images of subsurface articular cartilage degeneration, but its structural sensitivity has been limited to the micron scale. Most biological processes however, including early changes in OA occur at the nano-scale or cellular level. Nano-sensitive (nsOCT) is a new and emerging imaging modality capable of detecting structural changes in tissues at the nano-scale. The aim of this study was to investigate the capacity of nsOCT to detect structural changes in mesenchymal stem cell (MSC) chondrogenic pellets treated with inflammatory mediators as an ex vivo model of OA. Methods: An ex vivo experimental model of OA using MSC chondrogenic pellets with or without treatment, with inflammatory insults was developed. Four MSC cell pellets were prepared: (1) MSCs differentiated to the chondrogenic lineage via induction with 10ng/ml transforming growth factor (TGF)-ß3, (2) control/undifferentiated MSCs, (3) MSCs differentiated to the chondrogenic lineage with the addition of lipopolysaccharide (LPS) and palmitic acid (PA) for the last 24 hours (to mimic response to an acute inflammatory insult) and (4) MSCs differentiated to the chondrogenic lineage with the addition of alarmins (S100A8/A9) every 48 hours plus the addition of LPS and PA on the final day (to mimic chronic changes associated with OA). Conventional OCT, nsOCT, SESF microscopy (an additional imaging technique capable of monitoring nano-scale structural changes) and histological analysis was performed on MSC pellets. For OCT, nsOCT and SESF imaging, cell pellets were harvested and fixed before imaging. Histological analysis and staining with safranin-O of corresponding cell pellets was also performed to validate the nsOCT findings. Results: nsOCT imaging and SESF microscopy, but not conventional OCT imaging, were capable of detecting structural changes between groups. nsOCT plots are presented as enface images, formed as colour maps of the dominant spatial period of the structure. These colour maps characterise the structure, with blue structures indicating a smaller mean spatial period, and red structures indicating a larger mean spatial period of the sample. Using nsOCT imaging and SESF microscopy, the experimental groups were correlated to each other, with chondrogenic cell pellets and pellets treated with LPS/PA (acute mimic) demonstrating a similar structure, and undifferentiated control pellets and pellets treated with LPS/PA/alarmins (chronic mimic) demonstrating a similar structure. Chondrogenic pellets treated with TGF-ß3 demonstrated the largest spatial periods or dominant structures of all samples. This result indicated that MSCs underwent chondrogenic differentiation, the process by which terminal cartilage is formed from condensed mesenchyme tissue. This process consists of mesenchymal condensation followed by subsequent differentiation into chondrocytes through defined stages. The result is the deposition of a collagen II- and aggrecan-rich proteoglycan matrix that has structural organization, which was detectable using nsOCT imaging of these samples. Histological analysis confirmed these findings, as highly positive metachromasia and proteoglycan accumulation was evident during analysis. LPS/PA-treated samples had the second largest mean spatial period. This treatment was developed to mimic acute OA, and so only a minor reduction in structural integrity of these samples was observed. Undifferentiated cell pellets were found to have the next highest mean spatial period. This result was expected as although not stimulated to differentiate and synthesize a proteoglycan matrix due to the absence of TGF-ß3, these pellets maintained some structural integrity, which was detectable using nano-sensitive imaging. Finally, LPS/PA/alarmin-treated pellets demonstrated the lowest mean spatial period and dominant size of all groups, indicative of significant disruption to the internal structures of these samples. This structural disruption was also validated by histological analysis, where depletion and loss of proteoglycan matrix at the periphery of cell pellets was evident following staining with safranin-O. Conclusions: nsOCT and SESF microscopy were capable of detecting nano-scale structural changes in stem cell pellets following treatment with inflammatory insults as an ex vivo model for OA, and these nano-scale structural changes were validated with histological analysis. This study has demonstrated the capacity of nsOCT to detect nano-sensitive changes in tissue and the potential of this technique for diagnostic purposes in OA and various other conditions. Future work will assess the capacity of nsOCT to differentiate between healthy articular cartilage and osteoarthritic cartilage by performing nsOCT imaging of human cartilage explants.