One of the most important characteristic of knee osteoarthritis (OA) is the joint space (JS) width narrowing. Measurements are usually performed on two dimensional (2D) X‐rays. We propose and validate a new method to assess the 3D joint space at the medial knee compartment using high resolution peripheral computed tomography images. A semi‐automated method was developed to obtain a distance 3D map between femur an tibia with the following parameters: volume, minimum, maximum, mean, standard deviation, median, asymmetry, and entropy. We analyzed 71 knee specimens (mean age: 85 years), radiographs were performed for the Kellgren Lawrence (KL) score grading. In a subgroup of 41 specimens, the histopathological Outerbridge and meniscal classifications were performed and then cores were harvested from the tibial plateau in three different positions (posterior, central, and peripheral) and imaged at 10 µm of resolution to measure the cartilage thickness. Minimum, maximum, mean, and median were statistically lower and entropy higher between knee specimens classified as KL = 0 and KL = 3–4. Gr1 and 2 were statistically different from Gr3‐4 for minimum, asymmetry, entropy using the Outerbridge classification and Gr1 was statistically different from Gr3–4 using the meniscal classification. Asymmetry, minimum, mean, median and entropy were significantly correlated with cartilage thickness. Parameters extracted from a 3D map of the medial joint space indicate local variations of JS and are related to local measurements of tibial cartilage thickness, and could be consequently useful to identify early OA. © 2018 The Authors. Journal of Orthopaedic Research® Published by Wiley Periodicals, Inc. on behalf of Orthopaedic Research Society. J Orthop Res 36:2380–2391, 2018.

No abstract

No abstract

Abstract. Skeletonization is a morphological operation that summarizes an object by its median lines while preserving the initial image topology. It provides features used in biometric for the matching process, as well as medical imaging for quantification of the bone microarchitecture. We develop a solution for the extraction of structural and morphometric features useful in biometric, character recognition and medical imaging. It aims at storing object descriptors in a re-usable and hierarchical format. We propose graph data structures to identify skeleton nodes and branches, link them and store their corresponding features. This graph structure allows us to generate CSV files for high level analysis and to propose a pruning method that removes spurious branches regarding their length and mean gray level. We illustrate manipulations of the skeleton graph structure on medical image dedicated to bone microarchitecture characterization.

No abstract