Method for Segmentation of Knee Articular Cartilages Based on Contrast-Enhanced CT Images
Segmentation of contrast-enhanced computed tomography (CECT) images enables quantitative evaluation of morphology of articular cartilage as well as the significance of the lesions. Unfortunately, automatic segmentation methods for CECT images are currently lacking. Here, we introduce a semiautomated technique to segment articular cartilage from in vivo CECT images of human knee. The segmented cartilage geometries of nine knee joints, imaged using a clinical CT-scanner with an intra-articular contrast agent, were compared with manual segmentations from CT and magnetic resonance (MR) images. The Dice similarity coefficients (DSCs) between semiautomatic and manual CT segmentations were 0.79–0.83 and sensitivity and specificity values were also high (0.76–0.86). When comparing semiautomatic and manual CT segmentations, mean cartilage thicknesses agreed well (intraclass correlation coefficient = 0.85–0.93); the difference in thickness (mean ± SD) was 0.27 ± 0.03 mm. Differences in DSC, when MR segmentations were compared with manual and semiautomated CT segmentations, were statistically insignificant. Similarly, differences in volume were not statistically significant between manual and semiautomatic CT segmentations. Semiautomation decreased the segmentation time from 450 ± 190 to 42 ± 10 min per joint. The results reveal that the proposed technique is fast and reliable for segmentation of cartilage. Importantly, this is the first study presenting semiautomated segmentation of cartilage from CECT images of human knee joint with minimal user interaction.
KeywordsAutomation Contrast agent Knee joint Musculoskeletal Imaging
The authors acknowledge the Research Committee of the Kuopio University Hospital Catchment Area for the State Research Funding (Projects 5041746, 5041757, and 5203101). Study is also supported by Doctoral Program in Science, Technology and Computing (SCITECO, University of Eastern Finland), Finnish Cultural Foundation, and Academy of Finland (Projects 269315 and 307932).
Conflict of interest
Authors declare no conflicts of interest.
- 3.Buckwalter, J. A., J. L. Marsh, T. Brown, A. Amendola, and J. A. Martin. Articular cartilage injury. In: Principles of Tissue Engineering, edited by R. Lanza, R. Langer, and J. P. Vacanti. Elsevier, 2014, pp. 1253–1266.Google Scholar
- 7.Favre, J., S. F. Scanlan, J. C. Erhart-Hledik, K. Blazek, and T. P. Andriacchi. Patterns of femoral cartilage thickness are different in asymptomatic and osteoarthritic knees and can be used to detect disease-related differences between samples. J. Biomech. Eng. 135:101002, 2013.CrossRefGoogle Scholar
- 12.Guermazi, A., D. Hayashi, F. W. Roemer, J. Niu, E. K. Quinn, M. D. Crema, M. C. Nevitt, J. Torner, C. E. Lewis, and D. T. Felson. Partial- and full-thickness focal cartilage defects equally contribute to development of new cartilage damage in knee osteoarthritis—the Multicenter Osteoarthritis Study. Arthritis Rheumatol. 69:560–564, 2016.CrossRefGoogle Scholar
- 14.Kokkonen, H. T., J. Mäkelä, K. A. M. Kulmala, L. Rieppo, J. S. Jurvelin, V. Tiitu, H. M. Karjalainen, R. K. Korhonen, V. Kovanen, and J. Töyräs. Computed tomography detects changes in contrast agent diffusion after collagen cross-linking typical to natural aging of articular cartilage. Osteoarthr. Cartil. 19:1190–1198, 2011.CrossRefGoogle Scholar
- 17.Kroon, D. J. Finite Iterative Closest Point. Natick: Mathworks, 2009.Google Scholar
- 18.Kulmala, K. A. M., H. M. Karjalainen, H. T. Kokkonen, V. Tiitu, V. Kovanen, M. J. Lammi, J. S. Jurvelin, R. K. Korhonen, and J. Töyräs. Diffusion of ionic and non-ionic contrast agents in articular cartilage with increased cross-linking-contribution of steric and electrostatic effects. Med. Eng. Phys. 35:1415–1420, 2013.CrossRefGoogle Scholar
- 19.Lakin, B. A., H. Patel, C. Holland, J. D. Freedman, J. S. Shelofsky, B. D. Snyder, K. S. Stok, and M. W. Grinstaff. Contrast-enhanced CT using a cationic contrast agent enables non-destructive assessment of the biochemical and biomechanical properties of mouse tibial plateau cartilage. J. Orthop. Res. 34:1130–1138, 2016.CrossRefGoogle Scholar
- 21.Li, H., R. W. Sumner, and M. Pauly. Global correspondence optimization for non-rigid registration of depth scans. Eurograph. Symp. Geom. Process. 27:1421–1430, 2008.Google Scholar
- 25.Moshtagh, P. R. R., B. Pouran, J. van Tiel, J. Rauker, M. R. R. Zuiddam, V. Arbabi, N. M. M. Korthagen, H. Weinans, and A. A. A. Zadpoor. Micro- and nano-mechanics of osteoarthritic cartilage: the effects of tonicity and disease severity. J. Mech. Behav. Biomed. Mater. 59:561–571, 2016.CrossRefGoogle Scholar
- 26.Myller, K. A. H., M. J. Turunen, J. T. J. Honkanen, S. P. Väänänen, J. T. Iivarinen, J. Salo, J. S. Jurvelin, and J. Töyräs. In vivo contrast-enhanced cone beam CT provides quantitative information on articular cartilage and subchondral bone. Ann. Biomed. Eng. 45:811–818, 2017.CrossRefGoogle Scholar
- 37.Stewart, R. C., J. T. J. Honkanen, H. T. Kokkonen, V. Tiitu, S. Saarakkala, A. Joukainen, B. D. Snyder, J. S. Jurvelin, M. W. Grinstaff, and J. Töyräs. Contrast-enhanced computed tomography enables quantitative evaluation of tissue properties at intrajoint regions in cadaveric knee cartilage. Cartilage 8:391–399, 2017.CrossRefGoogle Scholar
- 45.van Tiel, J., M. Siebelt, J. H. Waarsing, T. M. Piscaer, M. van Straten, R. Booij, M. L. Dijkshoorn, G. J. Kleinrensink, J. A. N. Verhaar, G. P. Krestin, H. Weinans, and E. H. G. Oei. CT arthrography of the human knee to measure cartilage quality with low radiation dose. Osteoarthr. Cartil. 20:678–685, 2012.CrossRefGoogle Scholar