Abstract
We have developed a method for segmenting tibial and femoral medial cartilage in MR knee scans by combining two k Nearest Neighbors (kNN) classifiers for the cartilage classes with a rejection threshold for the background class. We show that with this threshold, two binary classifiers are sufficient compared to three binary classifiers in the traditional one-versus-all approach. We also show that the combination of binary classifiers produces better results than a kNN classifier that is trained to partition the voxels directly into three classes. The resulting sensitivity, specificity and Dice volume overlap of our method are 84.2%, 99.9% and 0.81 respectively. Compared to state-of-the-art segmentation methods, our method outperforms a fully automatic method and is comparable to a semi-automatic method.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Felson, D.T., Lawrence, R.C., Hochberg, M.C., McAlindon, T., Dieppe, P.A., Minor, M.A., Blair, S.N., Berman, B.M., Fries, J.F., Weinberger, M., Lorig, K.R., Jacobs, J.J., Goldberg, V.: Osteoarthritis: New insights, part 2: Treatment approaches. Annals of Internal Medicine 133, 726–737 (2000)
Graichen, H., Eisenhart-Rothe, R., Vogl, T., Englmeier, K.H., Eckstein, F.: Quantitative assessment of cartilage status in osteoarthritis by quantitative magnetic resonance imaging. Arthritis and Rheumatism 50, 811–816 (2004)
Pessis, E., Drape, J.L., Ravaud, P., Chevrot, A., Ayral, M.D.X.: Assessment of progression in knee osteoarthritis: results of a 1 year study comparing arthroscopy and mri. Osteoarthritis and Cartilage 11, 361–369 (2003)
Ejbjerg, B., Narvestad, E., Adn, H.S., Thomsen, S.J., Ostergaard, M.: Optimised, low cost, low field dedicated extremity mri is highly specific and sensitive for synovitis and bone erosions in rheumatoid arthritis wrist and finger joints: a comparison with conventional high-field mri and radiography. Annals of the Rheumatic Diseases 13 (2005)
Lynch, J.A., Zaim, S., Zhao, J., Stork, A., Peterfy, C.G., Genant, H.K.: Cartilage segmentation of 3d mri scans of the osteoarthritic knee combining user knowledge and active contours. In: SPIE, vol. 3979, pp. 925–935 (2000)
Solloway, S., Hutchinson, C., Vaterton, J., Taylor, C.: The use of active shape models for making thickness measurements of articular cartilage from mr images. Magnetic Resonance in Medicine 37, 943–952 (1997)
Grau, V., Mewes, A., Alcañiz, M., Kikinis, R., Warfield, S.: Improved watershed transform for medical image segmentation using prior information. IEEE Transactions on Medical Imaging 23 (2004)
Pakin, S.K., Tamez-Pena, J.G., Totterman, S., Parker, J.K.: Segmentation, surface extraction and thickness computation of articular cartilage. In: SPIE, vol. 4684, pp. 155–166 (2002)
Warfield, S.K., Kaus, M., Jolesz, F.A., Kikinis, R.: Adaptive, template moderated, spatially varying statistical classification. Medical Image Analysis, 43–55 (2000)
Dunn, T., Lu, Y., Jin, H., Ries, M., Majumdar, S.: T2 relaxation time of cartilage at mr imaging: comparison with severity of knee osteoarthritis. Radiology 232, 592–598 (2004)
Folkesson, J., Dam, E., Olsen, O.F., Pettersen, P., Christiansen, C.: Automatic segmentation of the articular cartilage in knee mri using a hierarchical multi-class classification scheme. In: Duncan, J.S., Gerig, G. (eds.) MICCAI 2005. LNCS, vol. 3749, pp. 327–334. Springer, Heidelberg (2005)
Kellgren, J., Lawrence, J.: Radiological assessment of osteo-arthrosis. Annals of the Rheumatic Diseases 16 (1957)
Arya, S., Mount, D.M., Netanyahu, N.S., Silverman, R., Wu, A.Y.: An optimal algorithm for approximate nearest neighbor searching in fixed dimensions. Journal of the ACM 45, 891–923 (1998)
Florack, L.: The Syntactical Structure of Scalar Images. PhD thesis, University of Utrecht (1993)
Mossman, D.: Three-way rocs. Med. Decis. Making 19, 78–79 (1999)
Edwards, D.C., Metz, C.E., Nishikawa, R.M.: The hypervolume under the roc hypersurface of near-guessing and near-perfect observers in n-class classification tasks. IEEE Transactions on Medical Imaging 24, 293–299 (2005)
Yeang, C.H., Ramaswamy, S., Tamayo, P., Mukherjee, S., Rifkin, R.M., Angelo, M., Reich, M., Lander, E., Mesirov, J., Golub, T.: Molecular classification of multiple tumor types. Bioinformatics 1 (2001)
Zou, K.H., Warfield, S.K., Bharatha, A., Tempany, C.M., Kaus, M.R., Haker, S.J., III, W.M.W., Jolesz, F.A., Kikinis, R.: Statistical validation of image segmentation quality based on a spatial overlap index. Acad Radiol 11 (2004)
Warfield, S.K., Zou, K.H., Wells, W.M.: Simultaneous truth and performance level estimation (staple). An algorithm for the validation of image segmentation 23 (2004)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Folkesson, J., Olsen, O.F., Pettersen, P., Dam, E., Christiansen, C. (2005). Combining Binary Classifiers for Automatic Cartilage Segmentation in Knee MRI. In: Liu, Y., Jiang, T., Zhang, C. (eds) Computer Vision for Biomedical Image Applications. CVBIA 2005. Lecture Notes in Computer Science, vol 3765. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11569541_24
Download citation
DOI: https://doi.org/10.1007/11569541_24
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-29411-5
Online ISBN: 978-3-540-32125-5
eBook Packages: Computer ScienceComputer Science (R0)