Fast identification & modelling of osteoarthritic holes in the human knee with contour interpolated radial basis functions

In this article, we propose a new method for the fast 3D reconstruction of simulated osteoarthritic holes in a human patella. Initially, clean patella slices were extracted from MRI scans of the knee using Region-Based Segmentation (region-growing) and Bounding Box techniques. Osteoarthritic (OA) holes were then simulated in the patella slices. Our contour interpolation/RBF (CI/RBF) method was then used to detect the hole automatically and reconstruct 3D models of both the patella and OA hole separately. The method presented here proves fast, reliable and efficient for reconstructing a 3D model of the patella from MRI images with an extremely low error of 0.33%.