Mahesh Nagarajan, Presenter: Nothing to Disclose

Markus B. Huber PhD, Abstract Co-Author: Nothing to Disclose

Paola Coan, Abstract Co-Author: Nothing to Disclose

Paul Claude Diemoz PhD, Abstract Co-Author: Nothing to Disclose

Christian Glaser MD, Abstract Co-Author: Nothing to Disclose

Axel Wismueller MD, PhD, Abstract Co-Author: Nothing to Disclose

Emmanuel Brun, Abstract Co-Author: Nothing to Disclose

Phase-contrast x-ray computed tomography (PCX-CT) is a recently developed novel imaging technique aimed at achieving soft-tissue contrast with micrometer scale resolution. This study proposes to quantitatively evaluate the performance of advanced texture analysis methods in characterizing chondrocyte patterns as observed in PCX-CT of human knee cartilage as healthy or osteoarhritic.

Five osteochondral cylinders (7 mm diameter, 2 osteoarthritic, 3 healthy) extracted from post-mortem human patellae were subject to PCX-CT at 26 keV (European Synchrotron Radiation Facility, Grenoble, France). 842 regions of interest (ROI) capturing chondrocyte patterns were then annotated in high resolution images (voxel size: 8 μm3) subsequently acquired from the radial zone of the cartilage matrix. Two texture analysis techniques - (1) Scaling Index Method (SIM), that estimates local scaling properties and (2) Gray Level Co-occurrence Matrix (GLCM), that computes second-order statistics, were used to extract features characterizing the chondrocyte patterns observed. Random sub-sampling cross-validation was utilized in optimizing a fuzzy k-Nearest Neighbor classifier; performance was measured using area under the Receiver-Operator Characteristic (ROC) curve (AUC) for each feature.

The best classification performance was observed with the SIM histogram and with GLCM correlation features (f3 and f13). With the experimental conditions used in this study, both SIM and GLCM achieved a high classification performance (AUC value of 0.94) in the task of distinguishing between health and osteoarthritic ROIs.

Phase Contrast CT has previously showcased its ability to visualize the internal architecture of the knee cartilage matrix. This study applies advanced texture analysis techniques to images acquired with PCX-CT to quantitatively evaluate their accuracy in distinguishing between healthy and osteoarthritic cartilage. Our results show that such features can capture differences in chondrocyte patterns annotated in the radial zone of knee cartilage matrix extracted from healthy and osteoarthritic specimens with high accuracy.

Computer-aided feature analysis can distinguish between osteoarthritic and healthy chondrocyte patterns in knee cartilage as seen on Phase Contrast CT imaging studies with micrometer scale resolution.

Nagarajan, M, Huber, M, Coan, P, Diemoz, P, Glaser, C, Wismueller, A, Brun, E, Phase Contrast Computed Tomography: Classification of Osteoarthritic and Healthy Chondrocyte Patterns in Human Patella Cartilage Using Local Scaling Property Estimation.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12033108.html