Quantification and visualization of MRI cartilage of the knee: A simplified approach

The estimation of articular cartilage of knee plays an important role in determining osteoarthritis (OA) level. The purpose of this study is to implement the segmentation, analysis and visualization techniques to characterize the knee cartilage in a simplified way. The segmentation technique used in this work is semi-automatic and based on Bezier splines and Canny edge detection. Cartilage edges are enhanced by using anisotropic diffusion to smooth the images. Shape-based interpolation is performed on the segmented cartilage in a simplified way for getting isotropic voxels. MRI registration is based on an approach which involves artificial matching of points at different slices. Analysis of the cartilage is then carried out by calculating its thickness and volume. Visualization of articular cartilage gives a supplementary tool to characterize it for quantification purpose.