Separation of healthy and early osteoarthritis by automatic quantification of cartilage homogeneity A.A. Qazi, M.Sc., J. Folkesson, M.Sc., P.C. Pettersen, M.D., M.A. Karsdal, Ph.D., C. Christiansen, M.D., Ph.D., E.B. Dam, Ph.D. Osteoarthritis and Cartilage Volume 15, Issue 10, Pages 1199-1206 (October 2007) DOI: 10.1016/j.joca.2007.03.016 Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions
Fig. 1 (a) Automatically segmented sagittal slice. (b) Cross section view of a segmented tibial medial cartilage sheet. Osteoarthritis and Cartilage 2007 15, 1199-1206DOI: (10.1016/j.joca.2007.03.016) Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions
Fig. 2 (a) Histogram of knee with highest entropy value in the data set (KL 0). (b) Histogram of knee with lowest entropy value (KL 3). (c) The same histogram in (a) but sorted by the number of bins. (d) The same histogram in (b) but sorted by the number of bins. The more diseased you are the lower will be the entropy. Osteoarthritis and Cartilage 2007 15, 1199-1206DOI: (10.1016/j.joca.2007.03.016) Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions
Fig. 3 Comparison of (a) JSW, (b) volume, (c) mean signal intensity, and (d) entropy as a function of the KL index. Entropy measure can separate healthy group from KL 1 much significantly when compared to the other measures. Moreover it can also separate healthy group from OA. Osteoarthritis and Cartilage 2007 15, 1199-1206DOI: (10.1016/j.joca.2007.03.016) Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions