Computer-aided classification of optical images for diagnosis of osteoarthritis in the finger joints

Article type: Research Article

Authors: Zhang, Jiang^; | Wang, James Z. | Yuan, Zhen | Sobel, Eric S. | Jiang, Huabei

Affiliations: Department of Biomedical Engineering, University of Florida, Gainesville, FL, USA | The Information Research Institute, Southwest Jiaotong University, Chengdu, China | School of Computing, Clemson University, Clemson, SC, USA | Division of Rheumatology, College of Medicine, University of Florida, Gainesville, FL, USA

Note: [] Corresponding author. Tel.: +1 352 273 9336; Fax: +1 352 273 9221; E-mail: hjiang@bme.ufl.edu

Abstract: This study presents a computer-aided classification method to distinguish osteoarthritis finger joints from healthy ones based on the functional images captured by x-ray guided diffuse optical tomography. Three imaging features, joint space width, optical absorption, and scattering coefficients, are employed to train a Least Squares Support Vector Machine (LS-SVM) classifier for osteoarthritis classification. The 10-fold validation results show that all osteoarthritis joints are clearly identified and all healthy joints are ruled out by the LS-SVM classifier. The best sensitivity, specificity, and overall accuracy of the classification by experienced technicians based on manual calculation of optical properties and visual examination of optical images are only 85%, 93%, and 90%, respectively. Therefore, our LS-SVM based computer-aided classification is a considerably improved method for osteoarthritis diagnosis.

Keywords: Osteoarthritis, finger joints, x-rays, diffuse optical tomography, least squares support vector machine, statistical analysis

DOI: 10.3233/XST-2011-0312

Journal: Journal of X-Ray Science and Technology, vol. 19, no. 4, pp. 531-544, 2011