Optimized High Resolution 3D Dense-U-Net Network for Brain and Spine Segmentation
Source abstract
Abstract
The 3D image segmentation is the process of partitioning a digital 3D volumes into multiple segments. This paper presents a fully automatic method for high resolution 3D volumetric segmentation of medical image data using modern supervised deep learning approach. We introduce 3D Dense-U-Net neural network architecture implementing densely connected layers. It has been optimized for graphic process unit accelerated high resolution image processing on currently available hardware (Nvidia GTX 1080ti). The method has been evaluated on MRI brain 3D volumetric dataset and CT thoracic scan dataset for spine segmentation. In contrast with many previous methods, our approach is capable of precise segmentation of the input image data in the original resolution, without any pre-processing of the input image. It can process image data in 3D and has achieved accuracy of 99.72% on MRI brain dataset, which outperformed results achieved by human expert. On lumbar and thoracic vertebrae CT dataset it has achieved the accuracy of 99.80%. The architecture proposed in this paper can also be easily applied to any task already using U-Net network as a segmentation algorithm to enhance its results. Complete source code was released online under open-source license.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Medical Imaging and Analysis · Medical Image Segmentation Techniques · Advanced Neural Network Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Telecommunications, Brno University of Technology, Brno 616 00, Czech Republic
Centre for Advanced Studies, Dr. A.P.J. Abdul Kalam Technical University, Lucknow 226031, India
Funding
Ministry of the Interior of the Czech Republic · National Sustainability Program · Ministry of Health of the Czech Republic
References (26)
Show source-provided references
1. Garcia-Garcia, A., Orts-Escolano, S., Oprea, S., Villena-Martinez, V., and Garcia-Rodriguez, J. (arXiv, 2017). A review on deep learning techniques applied to semantic segmentation, arXiv.10.1016/j.asoc.2018.05.018 ↗
2. Long, J., Shelhamer, E., and Darrell, T. (2015, January 7–12). Fully convolutional networks for semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.10.1109/CVPR.2015.7298965 ↗
3. Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. International Conference on Medical Image Computing And Computer-Assisted Intervention, Springer.10.1007/978-3-319-24574-4_28 ↗
4. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
5. Huang, G., Liu, Z., Van Der Maaten, L., and Weinberger, K.Q. (2017, January 21–26). Densely Connected Convolutional Networks. Proceedings of the Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.243 ↗
6. Jégou, S., Drozdzal, M., Vazquez, D., Romero, A., and Bengio, Y. (2017, January 21–26). The one hundred layers tiramisu: Fully convolutional densenets for semantic segmentation. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Honolulu, HI, USA.10.1109/CVPRW.2017.156 ↗
7. Milletari, F., Navab, N., and Ahmadi, S. (2016, January 25–28). V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation. Proceedings of the 2016 Fourth International Conference on 3D Vision (3DV), Stanford, CA, USA.10.1109/3DV.2016.79 ↗
8. Çiçek, Ö., Abdulkadir, A., Lienkamp, S.S., Brox, T., and Ronneberger, O. (arXiv, 2016). 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation, arXiv.10.1007/978-3-319-46723-8_49 ↗
9. Li, X., Chen, H., Qi, X., Dou, Q., Fu, C.W., and Heng, P.A. (arXiv, 2017). H-DenseUNet: Hybrid densely connected UNet for liver and liver tumor segmentation from CT volumes, arXiv.10.1109/TMI.2018.2845918 ↗
10. MRI segmentation of the human brain: challenges, methods, and applications
11. Uher, V., and Burget, R. (2012, January 3–4). Automatic 3D segmentation of human brain images using data-mining techniques. Proceedings of the 2012 35th International Conference on Telecommunications and Signal Processing (TSP), Prague, Czech Republic.10.1109/TSP.2012.6256362 ↗
12. Uher, V., Burget, R., Masek, J., and Dutta, M.K. (2013, January 2–4). 3D brain tissue selection and segmentation from MRI. Proceedings of the 2013 36th International Conference on Telecommunications and Signal Processing (TSP), Rome, Italy.10.1109/TSP.2013.6614057 ↗
13. Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm10.1109/42.906424 ↗
14. Deep learning for brain MRI segmentation: State of the art and future directions10.1007/s10278-017-9983-4 ↗
15. Kolařík, M., Burget, R., Uher, V., and Dutta, M.K. (2018, January 4–6). 3D Dense-U-Net for MRI Brain Tissue Segmentation. Proceedings of the IEEE 2018 41st International Conference on Telecommunications and Signal Processing (TSP), Athens, Greece.10.1109/TSP.2018.8441508 ↗
16. Lu, J.T., Pedemonte, S., Bizzo, B., Doyle, S., Andriole, K.P., Michalski, M.H., Gonzalez, R.G., and Pomerantz, S.R. (arXiv, 2018). DeepSPINE: Automated Lumbar Vertebral Segmentation, Disc-level Designation, and Spinal Stenosis Grading Using Deep Learning, arXiv.
17. Whitehead, W., Moran, S., Gaonkar, B., Macyszyn, L., and Iyer, S. (2018, January 4–7). A deep learning approach to spine segmentation using a feed-forward chain of pixel-wise convolutional networks. Proceedings of the 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), Washington, DC, USA.10.1109/ISBI.2018.8363709 ↗
18. Medical image file formats10.1007/s10278-013-9657-9 ↗
19. Yao, J., Burns, J.E., Munoz, H., and Summers, R.M. (2012). Detection of vertebral body fractures based on cortical shell unwrapping. International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer.10.1007/978-3-642-33454-2_63 ↗
20. Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion
21. Chollet, F. (2018, November 08). Keras. Available online: https://keras.io.
22. Noh, H., Hong, S., and Han, B. Learning deconvolution network for semantic segmentation. Proceedings of the Proceedings of the IEEE International Conference on Computer Vision, Washington, DC, USA, 7–13 December 2015.10.1109/ICCV.2015.178 ↗
23. Kolařík, M. (2018, November 08). mrkolarik/3D-brain-segmentation. Available online: https://github.com/mrkolarik/3D-brain-segmentation.
24. Jocić, M. (2018, May 01). Deep Learning Tutorial for Kaggle Ultrasound Nerve Segmentation Competition, Using Keras. Available online: https://github.com/jocicmarko/ultrasound-nerve-segmentation.
25. Statistical validation of image segmentation quality based on a spatial overlap index1: Scientific reports10.1016/S1076-6332(03)00671-8 ↗
26. Taha, A.A., and Hanbury, A. (2015). Metrics for evaluating 3D medical image segmentation: Analysis, selection, and tool. BMC Med. Imaging, 15.10.1186/s12880-015-0068-x ↗