ABSTRACT
Real-time cardiac magnetic resonance imaging (MRI) plays an increasingly important role in guiding various cardiac interventions. In order to provide better visual assistance, the cine MRI frames need to be segmented on-the-fly to avoid noticeable visual lag. In addition, considering reliability and patient data privacy, the computation is preferably done on local hardware. State-of-the-art MRI segmentation methods mostly focus on accuracy only, and can hardly be adopted for real-time application or on local hardware. In this work, we present the first hardware-aware multi-scale neural architecture search (NAS) framework for real-time 3D cardiac cine MRI segmentation. The proposed framework incorporates a latency regularization term into the loss function to handle realtime constraints, with the consideration of underlying hardware. In addition, the formulation is fully differentiable with respect to the architecture parameters, so that stochastic gradient descent (SGD) can be used for optimization to reduce the computation cost while maintaining optimization quality. Experimental results on ACDC MICCAI 2017 dataset demonstrate that our hardware-aware multi-scale NAS framework can reduce the latency by up to 3.5× and satisfy the real-time constraints, while still achieving competitive segmentation accuracy, compared with the state-of-the-art NAS segmentation framework.
- acdc challenge. 2017. https://www.creatis.insa-lyon.fr/Challenge/acdc/. (2017).Google Scholar
- Michelle Annett et al. 2014. How low should we go? Understanding the perception of latency while inking. In Proceedings of Graphics Interface 2014. 167--174.Google Scholar
- Olivier Bernard et al. 2018. Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE transactions on medical imaging 37, 11 (2018), 2514--2525.Google Scholar
- Song Bian, Xiaowei Xu, Weiwen Jiang, Yiyu Shi, and Takashi Sato. 2020. BUNET: Blind Medical Image Segmentation Based on Secure UNET. arXiv preprint arXiv:2007.06855 (2020).Google Scholar
- Han Cai, Ligeng Zhu, and Song Han. 2018. Proxylessnas: Direct neural architecture search on target task and hardware. arXiv preprint arXiv:1812.00332 (2018).Google Scholar
- Liang-Chieh Chen et al. 2018. Searching for efficient multi-scale architectures for dense image prediction. In Advances in neural information processing systems. 8699--8710.Google Scholar
- Yukun Ding, Jinglan Liu, Xiaowei Xu, Meiping Huang, Jian Zhuang, Jinjun Xiong, and Yiyu Shi. 2020. Uncertainty-Aware Training of Neural Networks for Selective Medical Image Segmentation. In Medical Imaging with Deep Learning.Google Scholar
- Nanqing Dong et al. 2019. Neural architecture search for adversarial medical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 828--836.Google Scholar
- Shunjie Dong, Jinlong Zhao, Maojun Zhang, Zhengxue Shi, Jianing Deng, Yiyu Shi, Mei Tian, and Cheng Zhuo. 2020. DeU-Net: Deformable U-Net for 3D Cardiac MRI Video Segmentation. arXiv preprint arXiv:2007.06341 (2020).Google Scholar
- Thomas Elsken et al. 2018. Neural architecture search: A survey. arXiv preprint arXiv:1808.05377 (2018).Google Scholar
- Thomas Gaspar et al. 2014. Three-dimensional real-time MRI-guided intracardiac catheter navigation. European heart journal 35, 9 (2014), 589--589.Google Scholar
- Peter W Iltis et al. 2015. High-speed real-time magnetic resonance imaging of fast tongue movements in elite horn players. Quantitative imaging in medicine and surgery 5, 3 (2015), 374.Google Scholar
- Fabian Isensee et al. 2017. Automatic cardiac disease assessment on cine-MRI via time-series segmentation and domain specific features. In International workshop on statistical atlases and computational models of the heart. Springer, 120--129.Google Scholar
- Fabian Isensee et al. 2018. nnu-net: Self-adapting framework for u-net-based medical image segmentation. arXiv preprint arXiv:1809.10486 (2018).Google Scholar
- Simon Jégou et al. 2017. The one hundred layers tiramisu: Fully convolutional densenets for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition workshops. 11--19.Google Scholar
- Weiwen Jiang et al. 2019. Accuracy vs. efficiency: Achieving both through fpgaimplementation aware neural architecture search. In Proceedings of the 56th Annual Design Automation Conference 2019. 1--6.Google Scholar
- Weiwen Jiang, Qiuwen Lou, Zheyu Yan, Lei Yang, Jingtong Hu, X Sharon Hu, and Yiyu Shi. 2020. Device-circuit-architecture co-exploration for computing-in-memory neural accelerators. IEEE Trans. Comput. (2020).Google ScholarCross Ref
- Weiwen Jiang, Lei Yang, Sakyasingha Dasgupta, Jingtong Hu, and Yiyu Shi. 2020. Standing on the shoulders of giants: Hardware and neural architecture co-search with hot start. arXiv preprint arXiv:2007.09087 (2020).Google Scholar
- Weiwen Jiang, Lei Yang, Edwin H-M Sha, Qingfeng Zhuge, Shouzhen Gu, Sakyasingha Dasgupta, Yiyu Shi, and Jingtong Hu. 2020. Hardware/Software co-exploration of neural architectures. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (2020).Google Scholar
- Sungwoong Kim et al. 2019. Scalable neural architecture search for 3d medical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 220--228.Google Scholar
- Yuhong Li et al. 2020. EDD: Efficient Differentiable DNN Architecture and Implementation Co-search for Embedded AI Solutions. arXiv preprint arXiv:2005.02563 (2020).Google Scholar
- Chenxi Liu et al. 2019. Auto-deeplab: Hierarchical neural architecture search for semantic image segmentation. 82--92.Google Scholar
- Hanxiao Liu, Karen Simonyan, and Yiming Yang. 2018. Darts: Differentiable architecture search. arXiv preprint arXiv:1806.09055 (2018).Google Scholar
- Zihao Liu et al. 2019. Machine vision guided 3d medical image compression for efficient transmission and accurate segmentation in the clouds. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 12687--12696.Google Scholar
- Qing Lu et al. 2019. On neural architecture search for resource-constrained hardware platforms. arXiv preprint arXiv:1911.00105 (2019).Google Scholar
- Elliot R McVeigh et al. 2006. Real-time interactive MRI-guided cardiac surgery: Aortic valve replacement using a direct apical approach. Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine 56, 5 (2006), 958--964.Google Scholar
- Aliasghar Mortazi et al. 2018. Automatically designing CNN architectures for medical image segmentation. In International Workshop on Machine Learning in Medical Imaging. Springer, 98--106.Google Scholar
- Toby Rogers et al. 2016. Transcatheter myocardial needle chemoablation during real-time magnetic resonance imaging: a new approach to ablation therapy for rhythm disorders. 9, 4 (2016), e003926.Google Scholar
- Sebastian Schaetz et al. 2017. Accelerated computing in magnetic resonance imaging: Real-time imaging using nonlinear inverse reconstruction. Computational and mathematical methods in medicine 2017 (2017).Google Scholar
- Mingxing Tan et al. 2019. Mnasnet: Platform-aware neural architecture search for mobile. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2820--2828.Google Scholar
- Tianchen Wang et al. 2019. MSU-Net: Multiscale Statistical U-Net for Real-time 3D Cardiac MRI Video Segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 614--622.Google Scholar
- Tianchen Wang, Xiaowei Xu, Jinjun Xiong, Qianjun Jia, Haiyun Yuan, Meiping Huang, Jian Zhuang, and Yiyu Shi. 2020. ICA-UNet: ICA Inspired Statistical UNet for Real-time 3D Cardiac Cine MRI Segmentation. arXiv preprint arXiv:2007.09455 (2020).Google Scholar
- Yu Weng et al. 2019. Nas-unet: Neural architecture search for medical image segmentation. IEEE Access 7 (2019), 44247--44257.Google ScholarCross Ref
- Bichen Wu, Xiaoliang Dai, Peizhao Zhang, Yanghan Wang, Fei Sun, Yiming Wu, Yuandong Tian, Peter Vajda, Yangqing Jia, and Kurt Keutzer. 2019. Fbnet: Hardware-aware efficient convnet design via differentiable neural architecture search. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 10734--10742.Google ScholarCross Ref
- Xiaowei Xu et al. 2018. Quantization of fully convolutional networks for accurate biomedical image segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition. 8300--8308.Google Scholar
- Xiaowei Xu et al. 2019. Whole heart and great vessel segmentation in congenital heart disease using deep neural networks and graph matching. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 477--485.Google Scholar
- Yuhui Xu et al. 2019. Pc-darts: Partial channel connections for memory-efficient architecture search. In International Conference on Learning Representations.Google Scholar
- Xingang Yan, Weiwen Jiang, Yiyu Shi, and Cheng Zhuo. 2020. MS-NAS: Multi-Scale Neural Architecture Search for Medical Image Segmentation. arXiv preprint arXiv:2007.06151 (2020).Google Scholar
- Lei Yang, Zheyu Yan, Meng Li, Hyoukjun Kwon, Liangzhen Lai, Tushar Krishna, Vikas Chandra, Weiwen Jiang, and Yiyu Shi. 2020. Co-Exploration of Neural Architectures and Heterogeneous ASIC Accelerator Designs Targeting Multiple Tasks. arXiv preprint arXiv:2002.04116 (2020).Google Scholar
- Qiao Zheng et al. 2018. 3-D consistent and robust segmentation of cardiac images by deep learning with spatial propagation. IEEE transactions on medical imaging 37, 9 (2018), 2137--2148.Google Scholar
- Clement Zotti et al. 2018. Convolutional neural network with shape prior applied to cardiac MRI segmentation. IEEE journal of biomedical and health informatics 23, 3 (2018), 1119--1128.Google Scholar
Recommendations
User intervention based segmentation of Myocardium In cardiac cine MRI images
WCI '15: Proceedings of the Third International Symposium on Women in Computing and InformaticsFor the past one and a half decade, Magnetic Resonance Imaging (MRI) has become a preferred imaging technique for examination of cardiac morphology and its functions in humans. The heart cavities segmentation in MRI is still a challenge to be resolved ...
Novelty towards Hybrid Segmentation of Coronary Artery in CT Cardiac Images
SNPD '08: Proceedings of the 2008 Ninth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed ComputingThe paper explains about the hybrid segmentation of coronary artery in CT cardiac images. The segmented volume is maximal intensity projected at different views to yield coronary angiograms showing the left anterior descending artery (LAD), right ...
Image registration and atlas-based segmentation of cardiac outflow velocity profiles
Cardiovascular disease is the leading cause of death worldwide and for this reason computer-based diagnosis of cardiac diseases is a very important task. In this article, a method for segmentation of aortic outflow velocity profiles from cardiac Doppler ...
Comments