Automatic Deep Feature Learning via Patch-Based Deep Belief Network for Vertebrae Segmentation in CT Images
Source abstract
Abstract
Precise automatic vertebra segmentation in computed tomography (CT) images is important for the quantitative analysis of vertebrae-related diseases but remains a challenging task due to high variation in spinal anatomy among patients. In this paper, we propose a deep learning approach for automatic CT vertebra segmentation named patch-based deep belief networks (PaDBNs). Our proposed PaDBN model automatically selects the features from image patches and then measures the differences between classes and investigates performance. The region of interest (ROI) is obtained from CT images. Unsupervised feature reduction contrastive divergence algorithm is applied for weight initialization, and the weights are optimized by layers in a supervised fine-tuning procedure. The discriminative learning features obtained from the steps above are used as input of a classifier to obtain the likelihood of the vertebrae. Experimental results demonstrate that the proposed PaDBN model can considerably reduce computational cost and produce an excellent performance in vertebra segmentation in terms of accuracy compared with state-of-the-art methods.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Medical Imaging and Analysis · Spinal Fractures and Fixation Techniques · Dental Radiography and Imaging
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China
Beijing Engineering Research Center of Mixed Reality and Advanced Display, School of Optics and Photonics, Beijing Institute of Technology, Beijing 100081, China
References (58)
Show source-provided references
1. Accurate Segmentation of Vertebral Bodies and Processes using Statistical Shape Decomposition and Conditional Models10.1109/TMI.2015.2396774 ↗
2. Automated model-based vertebra detection, identification, and segmentation in CT images10.1016/j.media.2009.02.004 ↗
3. Forsberg, D. (2015). Atlas-based registration for accurate segmentation of thoracic and lumbar vertebrae in CT data. Recent Advances in Computational Methods and Clinical Applications for Spine Imaging, Springer.10.1007/978-3-319-14148-0_5 ↗
4. Spine segmentation in medical images using manifold embeddings and higher-order MRFs10.1109/TMI.2013.2244903 ↗
5. Ibragimov, B., Korez, R., Likar, B., Pernuš, F., and Vrtovec, T. (2015). Interpolation-based detection of lumbar vertebrae in CT spine images. Recent Advances in Computational Methods and Clinical Applications for Spine Imaging, Springer.10.1007/978-3-319-14148-0_7 ↗
6. An improved level set method for vertebra CT image segmentation10.1186/1475-925X-12-48 ↗
7. Lootus, M., Kadir, T., and Zisserman, A. (2015). Automated radiological grading of spinal MRI. Recent Advances in Computational Methods and Clinical Applications for Spine Imaging, Springer.10.1007/978-3-319-14148-0_11 ↗
8. Dietterich, T.G. (2000). Ensemble methods in machine learning. International Workshop on Multiple Classifier Systems, Springer.10.1007/3-540-45014-9_1 ↗
9. Glocker, B., Feulner, J., Criminisi, A., Haynor, D.R., and Konukoglu, E. (2012). Automatic localization and identification of vertebrae in arbitrary field-of-view CT scans. International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer.10.1007/978-3-642-33454-2_73 ↗
10. Deep learning for automatic localization, identification, and segmentation of vertebral bodies in volumetric MR images
11. A new accurate and precise 3-D segmentation method for skeletal structures in volumetric CT data10.1109/TMI.2003.812265 ↗
12. Aslan, M.S., Ali, A., Rara, H., and Farag, A.A. (2010, January 26–29). An automated vertebra identification and segmentation in CT images. Proceedings of the 2010 IEEE International Conference on Image Processing, Hong Kong, China.10.1109/ICIP.2010.5651959 ↗
13. Li, H., and Wang, Z. (2006, January 17–18). A seepage flow model for vertebra CT image segmentation. Proceedings of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference, Shanghai, China.
14. Introducing willmore flow into level set segmentation of spinal vertebrae10.1109/TBME.2012.2225833 ↗
15. Li, Y., Liang, W., Tan, J., and Zhang, Y. (2015, January 7–9). A novel automatically initialized level set approach based on region correlation for lumbar vertebrae CT image segmentation. Proceedings of the 2015 IEEE International Symposium on Medical Measurements and Applications (MeMeA) Proceedings, Turin, Italy.10.1109/MeMeA.2015.7145215 ↗
16. Blumfield, A., and Blumfield, E. (2014). Automated Vertebral Body Image Segmentation for Medical Screening. (US8891848B2), Google Patents.
17. Yao, J., O’Connor, S.D., and Summers, R.M. (2006, January 6–9). Automated spinal column extraction and partitioning. Proceedings of the 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 2006, Arlington, VA, USA.
18. Using mathematical morphology for the anatomical labeling of vertebrae from 3D CT-scan images10.1016/j.compmedimag.2006.12.001 ↗
19. Hierarchical segmentation and identification of thoracic vertebra using learning-based edge detection and coarse-to-fine deformable model
20. Lumbar Spine Segmentation Using a Statistical Multi-Vertebrae Anatomical Shape + Pose Model10.1109/TMI.2013.2268424 ↗
21. A fully automatic vertebra segmentation method using 3D deformable fences10.1016/j.compmedimag.2009.02.006 ↗
22. Automatic inference of articulated spine models in CT images using high-order Markov Random Fields10.1016/j.media.2011.01.006 ↗
23. Shape representation for efficient landmark-based segmentation in 3-D10.1109/TMI.2013.2296976 ↗
24. Segmentation of lumbar vertebrae using part-based graphs and active appearance models
25. Parametric modelling and segmentation of vertebral bodies in 3D CT and MR spine images10.1088/0031-9155/56/23/011 ↗
26. Wang, Y., Yao, J., Roth, H.R., Burns, J.E., and Summers, R.M. (2015). Multi-atlas segmentation with joint label fusion of osteoporotic vertebral compression fractures on CT. International Workshop on Computational Methods and Clinical Applications for Spine Imaging, Springer.10.1007/978-3-319-41827-8_7 ↗
27. Learning-based vertebra detection and iterative normalized-cut segmentation for spinal MRI10.1109/TMI.2009.2023362 ↗
28. Representation learning: A review and new perspectives10.1109/TPAMI.2013.50 ↗
29. Learning hierarchical features for scene labeling10.1109/TPAMI.2012.231 ↗
30. The segmentation of the left ventricle of the heart from ultrasound data using deep learning architectures and derivative-based search methods10.1109/TIP.2011.2169273 ↗
31. A fast learning algorithm for deep belief nets10.1162/neco.2006.18.7.1527 ↗
32. Randomized algorithms for nonlinear system identification with deep learning modification
33. Patch-based segmentation using expert priors: Application to hippocampus and ventricle segmentation10.1016/j.neuroimage.2010.09.018 ↗
34. A supervised patch-based approach for human brain labeling10.1109/TMI.2011.2156806 ↗
35. Example-based super-resolution10.1109/38.988747 ↗
36. Efros, A.A., and Freeman, W.T. (2001, January 12–17). Image quilting for texture synthesis and transfer. Proceedings of the 28th Annual Conference on Computer Graphics and Interactive Techniques, Los Angeles, CA, USA.10.1145/383259.383296 ↗
37. Buades, A., Coll, B., and Morel, J.-M. (2005, January 20–25). A non-local algorithm for image denoising. Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), San Diego, CA, USA.
38. Cordier, N., Menze, B., Delingette, H., and Ayache, N. (2013). Patch-based segmentation of brain tissues. MICCAI Challenge on Multimodal Brain Tumor Segmentation, IEEE.
39. Wang, Z., Donoghue, C., and Rueckert, D. (2013). Patch-based segmentation without registration: Application to knee MRI. International Workshop on Machine Learning in Medical Imaging, Springer.10.1007/978-3-319-02267-3_13 ↗
40. Patch-based corner detection for cervical vertebrae in X-ray images10.1016/j.image.2017.04.002 ↗
41. Fully automatic cervical vertebrae segmentation framework for X-ray images10.1016/j.cmpb.2018.01.006 ↗
42. Ahmad, M., Yang, J., Ai, D., Qadri, S.F., and Wang, Y. (2017). Deep-stacked auto encoder for liver segmentation. Chinese Conference on Image and Graphics Technologies, Springer.10.1007/978-981-10-7389-2_24 ↗
43. Iterative convolutional neural networks for automatic vertebra identification and segmentation in CT images
44. Qadri, S.F., Ahmad, M., Ai, D., Yang, J., and Wang, Y. (2018). Deep Belief Network Based Vertebra Segmentation for CT Images. Chinese Conference on Image and Graphics Technologies, Springer.10.1007/978-981-13-1702-6_53 ↗
45. A multi-center milestone study of clinical vertebral CT segmentation10.1016/j.compmedimag.2015.12.006 ↗
46. (2018, October 11). Available online: Index @ Spineweb.Digitalimaginggroup.Ca.
47. Automatic localization of vertebrae based on convolutional neural networks
48. Using deep belief network modelling to characterize differences in brain morphometry in schizophrenia10.1038/srep38897 ↗
49. Medical image analysis using wavelet transform and deep belief networks10.1016/j.eswa.2017.05.073 ↗
50. Li, J., Xi, B., Li, Y., Du, Q., and Wang, K. (2018). Hyperspectral classification based on texture feature enhancement and deep belief networks. Remote Sens., 10.10.3390/rs10030396 ↗
51. Kim, Y., Lee, H., and Provost, E.M. (2013, January 26–31). Deep learning for robust feature generation in audiovisual emotion recognition. Proceedings of the 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver, BC, Canada.10.1109/ICASSP.2013.6638346 ↗
52. Training restricted Boltzmann machines: An introduction10.1016/j.patcog.2013.05.025 ↗
53. Random search for hyper-parameter optimization
54. Assessing the accuracy of prediction algorithms for classification: An overview10.1093/bioinformatics/16.5.412 ↗
55. Measures of the amount of ecologic association between species10.2307/1932409 ↗
56. Étude comparative de la distribution florale dans une portion des Alpes et des Jura
57. Seitel, A., Rasoulian, A., Rohling, R., and Abolmaesumi, P. (2015). Lumbar and Thoracic Spine Segmentation Using a Statistical Multi-object Shape + Pose Model. Recent Advances in Computational Methods and Clinical Applications for Spine Imaging, Springer.10.1007/978-3-319-14148-0_19 ↗
58. Sekuboyina, A., Kukačka, J., Kirschke, J.S., Menze, B.H., and Valentinitsch, A. (2017). Attention-driven deep learning for pathological spine segmentation. International Workshop and Challenge on Computational Methods and Clinical Applications in Musculoskeletal Imaging, Springer.10.1007/978-3-319-74113-0_10 ↗