Abstract
We describe a system for fully automatic vertebra localisation and segmentation in 3D CT volumes containing arbitrary regions of the spine, with the aim of detecting osteoporotic fractures. To avoid the difficulties of high-resolution manual annotation on overlapping structures in 3D, the system consists of several 2D operations. First, a Random Forest regressor is used to localise the spinal midplane in a coronal maximum intensity projection. A 2D sagittal image showing the midplane is then produced. A second set of regressors are used to localise each vertebral body in this image. Finally, a Random Forest Regression Voting Constrained Local Model is used to segment each detected vertebra.
The system was evaluated on 402 CT volumes. 83% of vertebrae between T4 and L4 were detected and, of these, 97% were segmented with a mean error of less than or equal to \(1\,mm\). A simple classifier was applied to perform a fracture/non-fracture classification for each image, achieving 69% recall at 70% precision.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
The remainder of the evaluation was repeated with \(D_t=\pm 10\,mm\), but this produced no improvements in the accuracy of subsequent stages, and the results are not reported here.
References
Adams, J.E.: Opportunistic identification of vertebral fractures. J. Clin. Densitometry 19(1), 54–62 (2016)
Breiman, L.: Random forests. Mach. Learn. 45, 5–32 (2001)
Bromiley, P.A., Adams, J.E., Cootes, T.F.: Localisation of vertebrae on DXA images using constrained local models with random forest regression voting. In: Yao, J., Glocker, B., Klinder, T., Li, S. (eds.) Recent Advances in Computational Methods and Clinical Applications for Spine Imaging. Lecture Notes in Computational Vision and Biomechanics, vol. 20, pp. 156–172. Springer, Cham (2015)
Bromiley, P.A., Adams, J.E., Cootes, T.F.: Automatic localisation of vertebrae in DXA images using random forest regression voting. In: Vrtovec, T., Yao, J., Glocker, B., Klinder, T., Frangi, A., Zheng, G., Li, S. (eds.) CSI 2015. LNCS, vol. 9402, pp. 38–51. Springer, Cham (2016). doi:10.1007/978-3-319-41827-8_4
Cootes, T.F., Edwards, G.J., Taylor, C.J.: Active appearance models. IEEE TPAMI 23, 681–685 (2001)
Cootes, T.F., Taylor, C.J., Cooper, D.H., Graham, J.: Active shape models - their training and application. Comput. Vis. Image Und. 61(1), 38–59 (1995)
Cristinacce, D., Cootes, T.: Automatic feature localisation with constrained local models. Pattern Recogn. 41(10), 3054–3067 (2008)
Cummings, S.R., Melton, J.: Epidemiology and outcomes of osteoporotic fractures. Lancet 359(9319), 1761–1767 (2002)
Delmas, P.D., van de Langerijt, L., Watts, N.B., Eastell, R., Genant, H.K., Grauer, A., Cahall, D.L.: Underdiagnosis of vertebral fractures is a worldwide problem: the IMPACT study. J. Bone Miner. Res. 20(4), 557–563 (2005)
Genant, H.K., Wu, C.Y., Kuijk, C.V., Nevitt, M.C.: Vertebral fracture assessment using a semi-quantitative technique. J. Bone Miner. Res. 8(9), 1137–1148 (1993)
Lindner, C., Bromiley, P.A., Ionita, M., Cootes, T.F.: Robust and accurate shape model matching using random forest regression-voting. IEEE TPAMI 37(9), 1862–1874 (2015)
Operational Information for Commissioning: Diagnostic imaging dataset statistical release. Technical report, NHS, UK, May 2016. www.england.nhs.uk/statistics/wp-content/uploads/sites/2/2015/08/Provisional-Monthly-Diagnostic-Imaging-Dataset-Statistics-2016-05-19.pdf
Rachner, T.D., Khosla, S., Hofbauer, L.C.: Osteoporosis: now and the future. Lancet 377(9773), 1276–1287 (2011)
Viola, P., Jones, M.: Rapid object detection using a boosted cascade of simple features. In: Proceeding of CVPR, pp. 511–518. IEEE Computer Society (2001)
Williams, A.L., Al-Busaidi, A., Sparrow, P.J., Adams, J.E., Whitehouse, R.W.: Under-reporting of osteoporotic vertebral fractures on computed tomography. Eur. J. Radiol. 69(1), 179–183 (2009)
Acknowledgment
This publication presents independent research supported by the Health Innovation Challenge Fund (grant no. HICF-R7-414/WT100936), a parallel funding partnership between the Department of Health and Wellcome Trust. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health or Wellcome Trust. The authors acknowledge the invaluable assistance of Mrs Chrissie Alsop, Mr Stephen Capener, Mrs Imelda Hodgkinson, Mr Michael Machin, and Mrs Sue Roberts, who performed the manual annotations.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing AG
About this paper
Cite this paper
Bromiley, P.A., Kariki, E.P., Adams, J.E., Cootes, T.F. (2016). Fully Automatic Localisation of Vertebrae in CT Images Using Random Forest Regression Voting. In: Yao, J., Vrtovec, T., Zheng, G., Frangi, A., Glocker, B., Li, S. (eds) Computational Methods and Clinical Applications for Spine Imaging. CSI 2016. Lecture Notes in Computer Science(), vol 10182. Springer, Cham. https://doi.org/10.1007/978-3-319-55050-3_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-55050-3_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-55049-7
Online ISBN: 978-3-319-55050-3
eBook Packages: Computer ScienceComputer Science (R0)