Abstract
Automatic and accurate semantic segmentation of pathological structures in medical images is challenging because of noisy disturbance, deformable shapes of pathology, and low contrast between soft tissues. Classical superpixel-based classification algorithms suffer from edge leakage due to complexity and heterogeneity inherent in medical images. Therefore, we propose a deep U-Net with superpixel region merging processing incorporated for edge enhancement to facilitate and optimize segmentation. Our approach combines three innovations: (1) different from deep learning--based image segmentation, the segmentation evolved from superpixel region merging via U-Net training getting rich semantic information, in addition to gray similarity; (2) a bilateral filtering module was adopted at the beginning of the network to eliminate external noise and enhance soft tissue contrast at edges of pathogy; and (3) a normalization layer was inserted after the convolutional layer at each feature scale, to prevent overfitting and increase the sensitivity to model parameters. This model was validated on lung CT, brain MR, and coronary CT datasets, respectively. Different superpixel methods and cross validation show the effectiveness of this architecture. The hyperparameter settings were empirically explored to achieve a good trade-off between the performance and efficiency, where a four-layer network achieves the best result in precision, recall, F-measure, and running speed. It was demonstrated that our method outperformed state-of-the-art networks, including FCN-16s, SegNet, PSPNet, DeepLabv3, and traditional U-Net, both quantitatively and qualitatively. Source code for the complete method is available at https://github.com/Leahnawho/Superpixel-network.
- Radhakrishna Achanta, Appu Shaji, Kevin Smith, Aurélien Lucchi, Pascal Fua, and Sabine Süsstrunk. 2012. SLIC superpixels compared to state-of-the-art superpixel methods. IEEE Transactions on Pattern Analysis and Machine Intelligence 34, 11 (2012), 2274--2282.Google ScholarDigital Library
- Vijay Badrinarayanan, Alex Kendall, and Roberto Cipolla. 2017. SegNet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 12 (2017), 2481--2495.Google ScholarCross Ref
- Robin Brügger, Christian F. Baumgartner, and Ender Konukoglu. 2019. A partially reversible U-net for memory-efficient volumetric image segmentation. In Proceedings of the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI’19). 429--437.Google ScholarCross Ref
- Mahaman Sani Chaibou, Pierre-Henri Conze, Karim Kalti, Basel Solaiman, and Mohamed Ali Mahjoub. 2018. Adaptive strategy for superpixel-based region-growing image segmentation. arXiv:1803.06541.Google Scholar
- Jiansheng Chen, Zhengqin Li, and Bo Huang. 2017. Linear spectral clustering superpixel. IEEE Transactions on Image Processing 26, 7 (2017), 3317--3330.Google ScholarDigital Library
- Liang-Chieh Chen, George Papandreou, Iasonas Kokkinos, Kevin Murphy, and Alan L. Yuille. 2018. DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence 40, 4 (2018), 834--848.Google ScholarCross Ref
- Liang-Chieh Chen, George Papandreou, Florian Schroff, and Hartwig Adam. 2017. Rethinking atrous convolution for semantic image segmentation. arXiv:1706.05587.Google Scholar
- Hao Dong, Guang Yang, Fangde Liu, Yuanhan Mo, and Yike Guo. 2017. Automatic brain tumor detection and segmentation using U-Net based fully convolutional networks. In Proceedings of the 21st Conferrence on Medical Image Understanding and Analysis (MIUA’17). 506--517.Google ScholarCross Ref
- Michal Drozdzal, Gabriel Chartrand, Eugene Vorontsov, Mahsa Shakeri, Lisa Di-Jorio, An Tang, Adriana Romero, Yoshua Bengio, Chris Pal, and Samuel Kadoury. 2018. Learning normalized inputs for iterative estimation in medical image segmentation. Medical Image Analysis 44 (2018), 1--13.Google ScholarCross Ref
- Robert M. Haralick and Linda G. Shapiro. 1985. Image segmentation techniques. Computer Vision, Graphics, and Image Processing 29, 1 (1985), 100--132.Google ScholarCross Ref
- Bulat Ibragimov, Robert Korez, Bostjan Likar, Franjo Pernus, Lei Xing, and Tomaz Vrtovec. 2017. Segmentation of pathological structures by landmark-assisted deformable models. IEEE Transactions on Medical Imaging 36, 7 (2017), 1457--1469.Google ScholarCross Ref
- Bulat Ibragimov and Lei Xing. 2016. Segmentation of organs-at-risks in head and neck CT images using convolutional neural networks. Medical Physics 44 (2016), 547--557.Google ScholarCross Ref
- Bo Li, Wiro J. Niessen, Stefan Klein, Marius de Groot, Mohammad Arfan Ikram, Meike W. Vernooij, and Esther E. Bron. 2019. A hybrid deep learning framework for integrated segmentation and registration: Evaluation on longitudinal white matter tract changes. In Proceedings of the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI’19). 645--653.Google Scholar
- Zun Li, Congyan Lang, Jiashi Feng, Yidong Li, Tao Wang, and Songhe Feng. 2019. Co-saliency detection with graph matching. ACM Transactions on Intelligent Systems and Technology 10, 3 (2019), Article 22, 22 pages.Google ScholarDigital Library
- Geert J. S. Litjens, Thijs Kooi, Babak Ehteshami Bejnordi, Arnaud Arindra Adiyoso Setio, Francesco Ciompi, Mohsen Ghafoorian, Jeroen A. W. M. van der Laak, Bram van Ginneken, and Clara I. Sánchez. 2017. A survey on deep learning in medical image analysis. Medical Image Analysis 42 (2017), 60--88.Google ScholarCross Ref
- Hong Liu, Baoxi Liu, Hao Zhang, Liang Li, Xin Qin, and Guijuan Zhang. 2018. Crowd evacuation simulation approach based on navigation knowledge and two-layer control mechanism. Information Sciences 436-437 (2018), 247--267.Google Scholar
- Hong Liu, Bin Xu, Dianjie Lu, and Guijuan Zhang. 2018. A path planning approach for crowd evacuation in buildings based on improved artificial bee colony algorithm. Applied Soft Computing 68 (2018), 360--376.Google ScholarDigital Library
- Ming-Yu Liu, Oncel Tuzel, Srikumar Ramalingam, and Rama Chellappa. 2011. Entropy rate superpixel segmentation. In Proceedings of the 24th IEEE Conference on Computer Vision and Pattern Recognition (CVPR’11). 2097--2104.Google ScholarDigital Library
- Yong-Jin Liu, Minjing Yu, Bing-Jun Li, and Ying He. 2018. Intrinsic manifold SLIC: A simple and efficient method for computing content-sensitive superpixels. IEEE Transactions on Pattern Analysis and Machine Intelligence 40, 3 (2018), 653--666.Google ScholarCross Ref
- Jonathan Long, Evan Shelhamer, and Trevor Darrell. 2015. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’15). 3431--3440.Google ScholarCross Ref
- Alastair Philip Moore, Simon Prince, Jonathan Warrell, Umar Mohammed, and Graham Jones. 2008. Superpixel lattices. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’08). 1--8.Google ScholarCross Ref
- Dong Nan, Duyan Bi, Shiping Ma, Xiaolong Lou, and Jiacheng Ni. 2014. An adaptive bilateral filtering method based on parameter estimation. Journal of Central South University 45, 11 (2014), 3840--3845.Google Scholar
- Xiao Pan, Yuanfeng Zhou, Zhonggui Chen, and Caiming Zhang. 2019. Texture relative superpixel generation with adaptive parameters. IEEE Transactions on Multimedia 21, 8 (2019), 1997--2011.Google ScholarCross Ref
- Sérgio Pereira, Adriano Pinto, Victor Alves, and Carlos A. Silva. 2016. Brain tumor segmentation using convolutional neural networks in MRI images. IEEE Transactions on Medical Imaging 35, 5 (2016), 1240--1251.Google ScholarCross Ref
- David M. W. Powers. 2008. Evaluation evaluation. In Proceedings of the 18th European Conference on Artificial Intelligence (ECAI’08). 843--844.Google Scholar
- Xiaofeng Ren and Jitendra Malik. 2003. Learning a classification model for segmentation. In Proceedings of the 9th International Conference on Computer Vision (ICCV’03). 10--17.Google ScholarCross Ref
- Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-Net: Convolutional networks for biomedical image segmentation. In Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI’15). 234--241.Google ScholarCross Ref
- Tim Salimans and Diederik P. Kingma. 2016. Weight normalization: A simple reparameterization to accelerate training of deep neural networks. In Advances in Neural Information Processing Systems (NIPS’16). 901--909.Google Scholar
- Jo Schlemper, Jose Caballero, Joseph V. Hajnal, Anthony N. Price, and Daniel Rueckert. 2018. A deep cascade of convolutional neural networks for dynamic MR image reconstruction. IEEE Transactions on Medical Imaging 37, 2 (2018), 491--503.Google ScholarCross Ref
- Jianbo Shi and Jitendra Malik. 2000. Normalized cuts and image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 22, 8 (2000), 888--905.Google ScholarDigital Library
- Brahim Ait Skourt, Abdelhamid El Hassani, and Aicha Majda. 2018. Lung CT image segmentation using deep neural networks. Procedia Computer Science 127 (2018), 109--113.Google ScholarDigital Library
- David Stutz, Alexander Hermans, and Bastian Leibe. 2018. Superpixels: An evaluation of the state-of-the-art. Computer Vision and Image Understanding 166 (2018), 1--27.Google ScholarDigital Library
- John E. Vargas-Muñoz, Ananda S. Chowdhury, Eduardo B. Alexandre, Felipe L. Galvão, Paulo A. Vechiatto Miranda, and Alexandre X. Falcão. 2019. An iterative spanning forest framework for superpixel segmentation. IEEE Transactions on Image Processing 28, 7 (2019), 3477--3489.Google ScholarDigital Library
- Rubén Vera-Rodríguez, Julian Fiérrez, and Aythami Morales. 2019. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications:23rd Iberoamerican Congress, CIARP 2018, Madrid, Spain, November 19-22, 2018, Proceedings. Lecture Notes in Computer Science, Vol. 11401. Springer.Google Scholar
- Lei Xing, Elizabeth A. Krupinski, and Jing Cai. 2018. Artificial intelligence will soon revolutionize medical physics research and practice. Medical Physics 45, 5 (2018), 1791--1793.Google ScholarCross Ref
- Yongxia Zhang, Xuemei Li, Xifeng Gao, and Caiming Zhang. 2017. A simple algorithm of superpixel segmentation with boundary constraint. IEEE Transactions on Circuits and Systems for Video Technology 27, 7 (2017), 1502--1514.Google ScholarDigital Library
- Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, and Jiaya Jia. 2017. Pyramid scene parsing network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’17). 6230--6239.Google ScholarCross Ref
- Chunbiao Zhu, Wenhao Zhang, Thomas H. Li, Shan Liu, and Ge Li. 2019. Exploiting the value of the center-dark channel prior for salient object detection. ACM Transactions on Intelligent Systems and Technology 10, 3 (2019), Article 32, 20 pages.Google ScholarDigital Library
Index Terms
- Superpixel Region Merging Based on Deep Network for Medical Image Segmentation
Recommendations
Medical Image Segmentation Using H-minima Transform and Region Merging Technique
FIT '11: Proceedings of the 2011 Frontiers of Information TechnologyWatershed transform is a commonly used image segmentation method. The main problem with this segmentation technique is that of its sensitivity to noise and other irregularities which leads to over-segmentation. In this paper the over-segmentation ...
Superpixel/voxel medical image segmentation algorithm based on the regional interlinked value
AbstractMedical image segmentation can effectively overcome human perception with strong personal limitations. The superpixel/voxel segmentation method has strong adaptability and computational efficiency. It can effectively separate the diseased tissue ...
Incorporating priors for medical image segmentation using a genetic algorithm
Medical image segmentation is typically performed manually by a physician to delineate gross tumor volumes for treatment planning and diagnosis. Manual segmentation is performed by medical experts using prior knowledge of organ shapes and locations but ...
Comments