Characterization of the Textural Features from Quantitative MRI for Determination of Cartilage Degeneration

Texture features derived from quantitative MRI maps of cartilage have attracted increasing attention from the osteoarthritis (OA) community in recent years. In this work, texture analysis was used on T2 maps and validated using histological analysis. Some texture features (autocorrelation, contrast and entropy) correlated with the Mankin score; autocorrelation also correlated with collagen orientation calculated from PLM images. The correlation between image-derived features with histological quality scores can be a decisive step towards monitoring of cartilage regeneration in-vivo and help to identify therapies that restore articular cartilage quantity and quality.