A comprehensive dataset of histopathology images, grades and patient demographics for human Osteoarthritis Cartilage

Osteoarthritis (OA) is a leading cause of disability in older adults and takes substantial toll at personal, economic and societal levels. There is inadequate comprehension of OA disease progression specifically during the early phases of OA. This knowledge is critical to understanding the heterogeneity in OA progression as well as enable development of targeted therapeutics at the start of the disease rather than end-stage. Histopathology of cartilage is a common method used to assess in situ state of cartilage tissue. The data presented in this article assesses the histopathological status of human cartilage specimens collected from 90 patients (n = 180). Each specimen was processed for histology and stained with hematoxylin and eosin (HE) and safranin O fast-green (SafO) for acquiring brightfield images to visualize changes in cartilage structure, cells, gycosaminoglycan content and tidemark integrity. The unstained sections were imaged using polarized light microscopy (PLM) to visualize changes in collagen organization and composition within the cartilage specimen. All the specimens were systematically graded by three scorers using established primary OA cartilage grading systems including Histological–Histochemical Grading System (HHGS), advanced Osteoarthritis Research Society International (OARSI) system and Polarized Light Microscopy (PLM) scoring system. These data can be used by the OA community as an educational resource to train new reviewers (scorers), it serves as a comprehensive image database for experienced OA community to review the wide spectrum of histopathological features presented by these mild to moderate OA specimens, to define different OA-subtypes, and to generate hypothesis on OA progression mechanisms. Finally, the high quality images can be used to develop machine learning algorithms for classification of OA, automated detection and segmentation of existing or new OA features that can serve as early OA histopathological indicators