Abstract
Osteoporosis induced fractures occur worldwide about every 3 s. Vertebral compression fractures are early signs of the disease and considered risk predictors for secondary osteoporotic fractures. We present a detection method to opportunistically screen spine-containing CT images for the presence of these vertebral fractures. Inspired by radiology practice, existing methods are based on 2D and 2.5D features but we present, to the best of our knowledge, the first method for detecting vertebral fractures in CT using automatically learned 3D feature maps. The presented method explicitly localizes these fractures allowing radiologists to interpret its results. We train a voxel-classification 3D Convolutional Neural Network (CNN) with a training database of 90 cases that has been semi-automatically generated using radiologist readings that are readily available in clinical practice. Our 3D method produces an Area Under the Curve (AUC) of 95% for patient-level fracture detection and an AUC of 93% for vertebra-level fracture detection in a five-fold cross-validation experiment.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Vertebrae are named T1 to T12 for thoracic, L1 to L5 for lumbar and S1–S2 for sacral vertebrae (with numbers increasing from top to bottom).
- 2.
Since our training database has only 11 negative cases, we stratified the random sampling to ensure that each fold has a minimum of two negative cases.
- 3.
The (False Positive Rate, True Positive Rate) values have been interpolated to plot a smoother curve.
References
Bar, A., Wolf, L., Amitai, O.B., Toledano, E., Elnekave, E.: Compression fractures detection on CT. In: SPIE Medical Imaging, pp. 1013440–1013440. International Society for Optics and Photonics (2017)
Bromiley, P.A., Kariki, E.P., Adams, J.E., Cootes, T.F.: 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.) CSI 2016. LNCS, vol. 10182, pp. 51–63. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-55050-3_5
Buckens, C.F., et al.: Intra and interobserver reliability and agreement of semiquantitative vertebral fracture assessment on chest computed tomography. PLoS ONE 8(8), e71204 (2013)
Burns, J.E., Yao, J., Summers, R.M.: Vertebral body compression fractures and bone density: automated detection and classification on CT images. Radiology 284, 788–797 (2017). https://doi.org/10.1148/radiol.2017162100
Genant, H.K., Wu, C.Y., van Kuijk, C., Nevitt, M.C.: Vertebral fracture assessment using a semiquantitative technique. J. Bone Miner. Res. 8(9), 1137–1148 (1993)
Glocker, B., Zikic, D., Konukoglu, E., Haynor, D.R., Criminisi, A.: Vertebrae localization in pathological spine CT via dense classification from sparse annotations. In: Mori, K., Sakuma, I., Sato, Y., Barillot, C., Navab, N. (eds.) MICCAI 2013. LNCS, vol. 8150, pp. 262–270. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-40763-5_33
Kamnitsas, K., et al.: Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med. Image Anal. 36, 61–78 (2017)
Koenig, M., Spindler, W., Rexilius, J., Jomier, J., Link, F., Peitgen, H.O.: Embedding VTK and ITK into a visual programming and rapid prototyping platform. In: Medical Imaging 2006: Visualization, Image-Guided Procedures, and Display, vol. 6141, p. 61412O. International Society for Optics and Photonics (2006)
Mader, A.O., et al.: Detection and localization of landmarks in the lower extremities using an automatically learned conditional random field. In: Cardoso, M.J., et al. (eds.) GRAIL/MFCA/MICGen -2017. LNCS, vol. 10551, pp. 64–75. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-67675-3_7
Provost, F.J., Fawcett, T., et al.: Analysis and visualization of classifier performance: comparison under imprecise class and cost distributions. In: KDD, vol. 97, pp. 43–48 (1997)
Tomita, N., Cheung, Y.Y., Hassanpour, S.: Deep neural networks for automatic detection of osteoporotic vertebral fractures on CT scans. Comput. Biol. Med. 98, 8–15 (2018)
Valentinitsch, A., et al.: Opportunistic osteoporosis screening in multi-detector CT images via local classification of textures. Osteoporos. Int. 30, 1275–1285 (2019)
Waterloo, S., et al.: Prevalence of vertebral fractures in women and men in the population-based Tromsø study. BMC Musculoskelet. Disord. 13(1), 3 (2012)
Yao, J., Burns, J.E., Wiese, T., Summers, R.M.: Quantitative vertebral compression fracture evaluation using a height compass. In: SPIE Medical Imaging, p. 83151X. International Society for Optics and Photonics (2012)
Acknowledgements
The authors thank the patients, investigators and their teams who took part in this study. The first author is grateful for the comments and feedback provided by Kasper Claes and the discussions on model evaluation with Roberto D’Ambrosio. This study was funded by UCB Pharma and Amgen Inc.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Nicolaes, J. et al. (2020). Detection of Vertebral Fractures in CT Using 3D Convolutional Neural Networks. In: Cai, Y., Wang, L., Audette, M., Zheng, G., Li, S. (eds) Computational Methods and Clinical Applications for Spine Imaging. CSI 2019. Lecture Notes in Computer Science(), vol 11963. Springer, Cham. https://doi.org/10.1007/978-3-030-39752-4_1
Download citation
DOI: https://doi.org/10.1007/978-3-030-39752-4_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-39751-7
Online ISBN: 978-3-030-39752-4
eBook Packages: Computer ScienceComputer Science (R0)
