Four articles were accepted by MICCAI2019

Four papers from MIRACLE research group were accepted by MICCAI2019, among which three papers received consistent positive reviews, and got early acceptance without rebuttal by MICCAI2019. The full name of MICCAI is International Conference on Medical Image Computing and Computer Assisted Intervention; it is a top international conference in the field of medical image analysis. The 22nd MICCAI conference will be held in Shenzhen, China in October 2019. The conference received more than 1,700 submissions, and the acceptance rate in previous years was usually less than 30%.

  1. Encoding CT Anatomy Knowledge for Unpaired Chest X-ray Image Decomposition (Zeju Li, Han Li, Hu Han, Gonglei Shi, Jiannan Wang, S. Kevin Zhou)
  2. Chest X-ray (CXR) offers a 2D projection of overlapped anatomies, and is widely used for clinical diagnosis. There is clinical evidence supporting that decomposing an X-ray image into different components (e.g., bone, lung and soft tissue) improves diagnostic value. We hereby propose a decomposition generative adversarial network (DecGAN) to anatomically decompose a CXR image but with unpaired data. We leverage the anatomy knowledge embedded in CT, which features a 3D volume with clearly visible anatomies. Our key idea is to embed CT priori decomposition knowledge into the latent space of unpaired CXR autoencoder. Specifically, we train DecGAN with a decomposition loss, adversarial losses, cycle-consistency losses and a mask loss to guarantee that the decomposed results of the latent space preserve realistic body structure. Extensive experiments demonstrate that DecGAN provides superior unsupervised CXR bone suppression results and the feasibility of modulating CXR components by latent space disentanglement. Furthermore, we illustrate the diagnostic value of DecGAN and demonstrate that it outperforms the state-of-the-art approaches in terms of predicting 11 out of 14 common lung diseases.

Fig. 1. Overview of DecGAN

2. 3D U2-Net: A 3D Universal U-Net for Multi-Domain Medical Image Segmentation (Chao Huang, Hu Han, Qingsong Yao, Shankuan Zhu, S. Kevin Zhou)

Fully convolutional neural networks like U-Net have been the state-of-art methods in medical image segmentation. Practically, a network is highly specialized and trained separately for each segmentation task. Instead of a collection of multiple models, it is highly desirable to learn a universal data representation for different tasks, ideally a single model with the addition of a minimal number of parameters to steer to each task. Inspired by the recent success of multi-domain learning in image classification, for the first time we explore a promising universal architecture that can handle multiple medical segmentation tasks, regardless of different organs and imaging modalities. Our 3D Universal U-Net (3D U2-Net) is built upon separable convolution, assuming that images from different domains have domain-specific spatial correlations which can be probed with channel-wise convolution while also share cross-channel correlations which can be modeled with pointwise convolution. We evaluate the 3D U2-Net on five organ segmentation datasets. Experimental results show that this universal network is capable of competing with traditional models in terms of segmentation accuracy, while requiring only 1% of the parameters. Additionally, we observe that the architecture can be easily and effectively adapted to a new domain without sacrificing performance in the domains used to learn the shared parameterization of the universal network.

Fig. 2. Overview of 3D U2-net

  1. Generative Mask Pyramid Network forCT/CBCT Metal Artifact Reduction with Joint Projection-Sinogram Correction (Haofu Liao, Wei-An Lin, Zhimin Huo, Levon Vogelsang, William J. Sehnert, S. Kevin Zhou, Jiebo Luo)
  2. Artifact Disentanglement Network for Unsupervised Metal Artifact Reduction (Haofu Liao, Wei-An Lin, Jianbo Yuan, S. Kevin Zhou, Jiebo Luo)

 

[1] Zeju Li, Han Li, Hu Han, Gonglei Shi, Jiannan Wang, S. Kevin Zhou. “Encoding CT Anatomy Knowledge for Unpaired Chest X-ray Image Decomposition,” to appear in MICCAI 2019.

[2] Chao Huang, Hu Han, Qingsong Yao,  Shankuan Zhu, S. Kevin Zhou. “3D U2-Net: A 3D Universal U-Net for Multi-Domain Medical Image Segmentation,” to appear in MICCAI 2019.

[3] Haofu Liao, Wei-An Lin, Zhimin Huo, Levon Vogelsang, William J. Sehnert, S. Kevin Zhou, Jiebo Luo. “Generative Mask Pyramid Network forCT/CBCT Metal Artifact Reduction with Joint Projection-Sinogram Correction.” to appear in MICCAI 2019.

[4] Haofu Liao, Wei-An Lin, Jianbo Yuan, S. Kevin Zhou, Jiebo Luo.” Artifact Disentanglement Network for Unsupervised Metal Artifact Reduction.” to appear in MICCAI 2019.