Syaifiil Anam, Eiji Uchino, Noriaki Suetake
A hand bones radiograph is the gold standard for Rheumatoid Arthritis (RA) diagnosis. RA is an inflammatory disease that attacks the joint cartilage which causes premature mortality, disability, and it compromises the quality of life. Early diagnosis and treatment of RA can carefully delay joint destruction, disease activity, and functional disability. To diagnose RA, the hand bones radiograph is to be taken and analyzed. Before the hand bones radiograph is analyzed, the first step is that the hand bones radiograph is carefully segmented. However the hand bones radiograph segmentation is an extremely exhausting and time consuming task for radiologists, not only because the hand radiograph has low quality and uneven illumination, but also because it is very complex. The precise segmentation is required during RA diagnosis. Therefore an automatic segmentation method of bones is required. To correct the illumination and enhance the hand bones radiograph, we will employ a new morphology operation that is combined with a set of image processing. After correcting the illumination and enhancing the hand bones radiograph, the hand bones radiograph is segmented by applying fractal analysis. After the experiments on particular sets of hand radiograph images, we found that the proposed method works better compared with the conventional segmentation methods used in our previous works. © 2014 IEEE.
Graduate School of Science and Engineering, Yamaguchi University, 1677-1 Yoshida, Yamaguchi, 753-8512, Japan; Department of Mathematics, University of Brawijaya, Malang, 65145, Indonesia; Fuzzy Logic Systems Institute, 680-41 Kawazu, Iizuka, 820-0067, Japan
