The complex shape of embryonic cartilage represents a true challenge for phenotyping and basic understanding of skeletal development. X-ray computed microtomography (μCT) enables inspecting relevant tissues in all three dimensions; however, most 3D models are still created by manual segmentation, which is a time-consuming and tedious task. In this work, we utilised a convolutional neural network (CNN) to automatically segment the most complex cartilaginous system represented by the developing nasal capsule. The main challenges of this task stem from the large size of the image data (over a thousand pixels in each dimension) and a relatively small training database, including genetically modified mouse embryos, where the phenotype of the analysed structures differs from the norm. We propose a CNN-based segmentation model optimised for the large image size that we trained using a unique manually annotated database. The segmentation model was able to segment the cartilaginous nasal capsule with a median accuracy of 84.44% (Dice coefficient). The time necessary for segmentation of new samples shortened from approximately 8 h needed for manual segmentation to mere 130 s per sample. This will greatly accelerate the throughput of μCT analysis of cartilaginous skeletal elements in animal models of developmental diseases.

Central European Institute of Technology Brno University of Technology Purkynova 123 Brno 61200 Czech Republic

Department of Physiology and Pharmacology Karolinska Institutet 17165 Stockholm Sweden

Max Planck Institute for Evolutionary Biology August Thienemann Str 2 24306 Ploen Germany

Medical University of Vienna Spitalgasse 23 1090 Vienna Austria

Zobrazit více v PubMed

Metscher BD. MicroCT for developmental biology: A versatile tool for high-contrast 3D imaging at histological resolutions. Dev. Dyn. 2009 doi: 10.1002/dvdy.21857. PubMed DOI

Kaucka M, et al. Oriented clonal cell dynamics enables accurate growth and shaping of vertebrate cartilage. Elife. 2017 doi: 10.7554/eLife.25902. PubMed DOI PMC

Kaucka M, et al. Signals from the brain and olfactory epithelium control shaping of the mammalian nasal capsule cartilage. Elife. 2018 doi: 10.7554/eLife.34465. PubMed DOI PMC

De Bakker, B. S. et al. An interactive three-dimensional digital atlas and quantitative database of human development. Science (80-. ). (2016). 10.1126/science.aag0053. PubMed

Powell KA, Wilson D. 3-dimensional imaging modalities for phenotyping genetically engineered mice. Vet. Pathol. 2012 doi: 10.1177/0300985811429814. PubMed DOI PMC

Karaplis, A. C. Embryonic Development of Bone and Regulation of Intramembranous and Endochondral Bone Formation. in Principles of Bone Biology, Two-Volume Set (2008). 10.1016/B978-0-12-373884-4.00025-2.

Vanhoenacker FM, Van Hul W, Gielen J, De Schepper AM. Congenital skeletal abnormalities: An introduction to the radiological semiology. Eur. J. Radiol. 2001 doi: 10.1016/S0720-048X(01)00398-9. PubMed DOI

Neben CL, Roberts RR, Dipple KM, Merrill AE, Klein OD. Modeling craniofacial and skeletal congenital birth defects to advance therapies. Hum. Mol. Genet. 2016 doi: 10.1093/hmg/ddw171. PubMed DOI PMC

Brommage R, Ohlsson C. High fidelity of mouse models mimicking human genetic skeletal disorders. Front. Endocrinol. 2020 doi: 10.3389/fendo.2019.00934. PubMed DOI PMC

McBratney-Owen B, Iseki S, Bamforth SD, Olsen BR, Morriss-Kay GM. Development and tissue origins of the mammalian cranial base. Dev. Biol. 2008 doi: 10.1016/j.ydbio.2008.07.016. PubMed DOI PMC

Hatt M, Parmar C, Qi J, El Naqa I. Machine (Deep) learning methods for image processing and radiomics. IEEE Trans. Radiat. Plasma Med. Sci. 2019 doi: 10.1109/trpms.2019.2899538. DOI

Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. Lect. Notes Comput. Sci. 2015 doi: 10.1007/978-3-319-24574-4_28. DOI

Dong H, Yang G, Liu F, Mo Y, Guo Y. Automatic brain tumor detection and segmentation using U-net based fully convolutional networks. Commun. Comput. Inf. Sci. 2017 doi: 10.1007/978-3-319-60964-5_44. DOI

Degenhardt K, Wright AC, Horng D, Padmanabhan A, Epstein JA. Rapid 3D phenotyping of cardiovascular development in mouse embryos by micro-CT with iodine staining. Circ. Cardiovasc. Imaging. 2010 doi: 10.1161/CIRCIMAGING.109.918482. PubMed DOI PMC

Pai VM, et al. Coronary artery wall imaging in mice using osmium tetroxide and micro-computed tomography (micro-CT) J. Anat. 2012 doi: 10.1111/j.1469-7580.2012.01483.x. PubMed DOI PMC

Kaiser J, et al. Use of micro computed-tomography and 3D printing for reverse engineering of mouse embryo nasal capsule. J. Instrum. 2016;11:C03006–C03006. doi: 10.1088/1748-0221/11/03/C03006. DOI

Norman B, Pedoia V, Majumdar S. Use of 2D U-net convolutional neural networks for automated cartilage and meniscus segmentation of knee MR imaging data to determine relaxometry and morphometry. Radiology. 2018 doi: 10.1148/radiol.2018172322. PubMed DOI PMC

Zhu Q, Du B, Turkbey B, Choyke PL, Yan P. Deeply-supervised CNN for prostate segmentation. Proc. Int. Joint Conf. Neural Netw. 2017 doi: 10.1109/IJCNN.2017.7965852. DOI

Rytky SJO, et al. Automated analysis of rabbit knee calcified cartilage morphology using micro-computed tomography and deep learning. J. Anat. 2021 doi: 10.1111/joa.13435. PubMed DOI PMC

He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. 2016 doi: 10.1109/CVPR.2016.90. DOI

Jia Deng et al. ImageNet: A large-scale hierarchical image database. in (2009). 10.1109/cvprw.2009.5206848.

Léger J, Leyssens L, De Vleeschouwer C, Kerckhofs G. Deep learning-based segmentation of mineralized cartilage and bone in high-resolution micro-CT images. Lect. Notes Comput. Vis. Biomech. 2020 doi: 10.1007/978-3-030-43195-2_12. PubMed DOI

Zheng H, et al. Cartilage segmentation in high-resolution 3d micro-ct images via uncertainty-guided self-training with very sparse annotation. Lect. Notes Comput. Sci. 2020 doi: 10.1007/978-3-030-59710-8_78. PubMed DOI PMC

Metscher BD. Micro CT for comparative morphology: Simple staining methods allow high-contrast 3D imaging of diverse non-mineralized animal tissues. BMC Physiol. 2009 doi: 10.1186/1472-6793-9-11. PubMed DOI PMC

Matula, J. et al. X-ray microtomography–based atlas of mouse cranial development. Gigascience10, (2021). PubMed PMC

Matula, J. et al. Supporting data for ‘X-ray microtomography-based atlas of mouse cranial development’. GigaScience Database (2021). PubMed PMC

Feldkamp LA, Davis LC, Kress JW. Practical cone-beam algorithm. J. Opt. Soc. Am. A. 1984 doi: 10.1364/josaa.1.000612. DOI

Khanna A, Londhe ND, Gupta S, Semwal A. A deep Residual U-Net convolutional neural network for automated lung segmentation in computed tomography images. Biocybern. Biomed. Eng. 2020 doi: 10.1016/j.bbe.2020.07.007. DOI

Klambauer, G., Unterthiner, T., Mayr, A. & Hochreiter, S. Self-normalizing neural networks. in Advances in Neural Information Processing Systems (2017).

LeCun, Y. A., Bottou, L., Orr, G. B. & Müller, K. R. Efficient backprop. Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics) (2012). 10.1007/978-3-642-35289-8_3.

Chollet, F. Keras: The Python Deep Learning library. Keras.Io (2015).

Martin A, Paul B, Jianmin C, Zhifeng C. TensorFlow: A system for large-scale machine learning. OSDI. 2016;16:265–283.

Harris CR, et al. Array programming with NumPy. Nature. 2020 doi: 10.1038/s41586-020-2649-2. PubMed DOI PMC

Van Der Walt S, et al. Scikit-image: Image processing in python. PeerJ. 2014 doi: 10.7717/peerj.453. PubMed DOI PMC

Kingma, D. P. & Ba, J. L. Adam: A method for stochastic optimization. in 3rd International Conference on Learning Representations, ICLR 2015—Conference Track Proceedings (2015).

Reddi, S. J., Kale, S. & Kumar, S. On the convergence of adam and beyond. arXiv (2019).

Milletari, F., Navab, N. & Ahmadi, S. A. V-Net: Fully convolutional neural networks for volumetric medical image segmentation. in Proceedings—2016 4th International Conference on 3D Vision, 3DV 2016 565–571 (2016). 10.1109/3DV.2016.79.

Zou KH, et al. Statistical validation of image segmentation quality based on a spatial overlap index. Acad. Radiol. 2004 doi: 10.1016/S1076-6332(03)00671-8. PubMed DOI PMC

Theiler K. The house mouse. House Mouse. 1989 doi: 10.1007/978-3-642-88418-4. DOI