Arttu Peuna1,2,3, Joonas Hekkala1,3, Marianne Haapea1,2, Jana Podlipska1,3, Simo Saarakkala1,2,3, Miika T Nieminen1,2,3, and Eveliina Lammentausta1,2
1Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland, 2Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland, 3Research group of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
Texture analysis methods based on gray level
co-occurrence matrices can be optimized to probe the spatial information from
knee MR T2 maps and of the changes caused by osteoarthritis. Curvature of
the cartilage surfaces and relatively low resolution in relation to cartilage
thickness set special requirements for texture analysis tools. Here we report
an optimized analysis tool with customized point-wise offset angle and endpoint
extrapolation for cartilage texture analysis. Method provides excellent
performance compared to traditional texture analysis implementation.