BCD-Net for Low-Dose CT Reconstruction: Acceleration, Convergence, and Generalization
Obtaining accurate and reliable images from low-dose computed tomography (CT) is challenging. Regression convolutional neural network (CNN) models that are learned from training data are increasingly gaining attention in low-dose CT reconstruction. This paper modifies the architecture of an iterative regression CNN, BCD-Net, for fast, stable, and accurate low-dose CT reconstruction, and presents the convergence property of the modified BCD-Net. Numerical results with phantom data show that applying faster numerical solvers to model-based image reconstruction (MBIR) modules of BCD-Net leads to faster and more accurate BCD-Net; BCD-Net significantly improves the reconstruction accuracy, compared to the state-of-the-art MBIR method using learned transforms; BCD-Net achieves better image quality, compared to a state-of-the-art iterative NN architecture, ADMM-Net. Numerical results with clinical data show that BCD-Net generalizes significantly better than a state-of-the-art deep (non-iterative) regression NN, FBPConvNet, that lacks MBIR modules.
This work is supported in part by NSFC 61501292 and NIH U01 EB018753. The authors thank GE Healthcare for supplying the clinical data. The authors thank Zhipeng Li for his help with debugging the codes.
- 4.Chun, I.Y., Fessler, J.A.: Deep BCD-net using identical encoding-decoding CNN structures for iterative image recovery. In: Proceedings of IEEE IVMSP Workshop, Zagori, Greece, pp. 1–5, June 2018Google Scholar
- 5.Chun, I.Y., Fessler, J.A.: Convolutional analysis operator learning: acceleration and convergence. IEEE Trans. Image Process. (2019, to appear). https://arxiv.org/abs/1802.05584
- 6.Chun, I.Y., Huang, Z., Lim, H., Fessler, J.A.: Momentum-Net: fast and convergent recurrent neural network for inverse problems (2019, submitted). http://arxiv.org/abs/1907.11818
- 9.Kingma, D.P., Ba, J.L.: Adam: a method for stochastic optimization. In: Proceedings of ICLR 2015, San Diego, CA, pp. 1–15, May 2015Google Scholar
- 14.Yang, Y., Sun, J., Li, H., Xu, Z.: Deep ADMM-Net for compressive sensing MRI. In: Proceedings of NIPS \(29\), Long Beach, CA, pp. 10–18, December 2016Google Scholar
- 15.Zheng, X., Chun, I.Y., Li, Z., Long, Y., Fessler, J.A.: Sparse-view X-ray CT reconstruction using \(\ell _1\) prior with learned transform. Submitted, February 2019. http://arxiv.org/abs/1711.00905