新闻动态：
2020年5月，我组研究员周少华老师成为NeurIPS 2020的领域主席。
• Medical Image Recognition, Segmentation and Parsing 开源下载。
• MONAI 平台上线， 周少华研究员受邀加入咨询委员会。
MONAI, 是一个基于pytorch的开源医学AI平台。
• 韩琥老师受邀担任国际期刊模式识别(Pattern Recognition, CCF-B类，IF: 5.898) 编委(Associate Editor)。
2020年4月，韩琥老师受邀担任国际期刊模式识别(Pattern Recognition, CCF-B类，IF: 5.898) 编委(Associate Editor)。
人才荟萃：
欢迎肖立副研究员加入研究组
肖立副研究员，清华大学数学物理基础科学专业本科，美国加州大学欧文分校生物医学工程系博士，国家优秀自费留学生奖学金获得者。在TMI，MICCAI，Science，JCTC等国际顶尖学术期刊和会议上发表论文20余篇， 其中第一／通讯作者论文11篇。
论文发表：
- J. Zhu, Y. Li, Y. Hu, K. Ma, S. Kevin Zhou, Y. Zheng: Rubik’s Cube+: A Self-supervised Feature Learning Framework for 3D Medical Image Analysis. to appear in Med. Ima. Anal.
- 本文提出一个代理任务，即，魔方+，来预训练三维神经网络。代理任务包括三种操作，即立方体排序、立方体旋转和立方体遮挡。代理任务将迫使网络从原始三维医学数据中学习平移和旋转不变特征，同时对数据的噪声有一定容忍性。相比于从头开始的训练策略，利用魔方+预训练再进行微调，可以显著提高三维神经网络在脑出血分类、脑瘤分割等任务上的准确性，而不需要使用额外的数据。
- H. Li, H. Han, and S. Kevin Zhou, Bounding maps for universal lesion detection.
- CT图像的Universal Lesion 检测在计算机辅助诊断系统中扮演著重要的角色。许多检测方法通过使用可能的anchors取得了不错检测的结果。然而，经验证据表明，anchor-based 方向的data-imbalance问题将导致高假阳性(FP)率。本文提出了一种box-to-map的方法，用三个在x、y、xy方向上的soft continuous maps来表示BBox。我们的方法嵌入到四种state-of-the-art two-stage anchor-based检测方法中，可以有效地提高检测精度(1.68% to 3.85% boost of sensitivity at 4FPs)。
- Q. Yao, Z. He, H. Han, and S. Kevin Zhou, Miss the point: Targeted adversarial attack on multiple landmark detection.
- 本文首次将对抗攻击引入关键点检测任务中，使用I-FGSM对U-Net（具有代表性，用于关键点检测）的heatmap和offset map两个分支同时攻击，实验证明我们可以将任意关键点的预测结果移动到任意位置上，这会对病人的疾病诊疗产生很大的威胁。进一步，我们通过自适应改变每个关键点的权重，让目标攻击更加快速和有效。
- Z. Huang, Y. Ding, G. Song, L. Wang, R. Geng, H. He, S. Du, X. Liu, Y. Tian, Y. Liang, S. Kevin Zhou, and J. Chen, BCData: A large-scale dataset and benchmark for cell detection and counting.
- Y. Lyu, W. Lin, H. Liao, J. Lu, and S. Kevin Zhou, Encoding metal mask projection for metal artifact reduction in computed tomography.
- W. Wang, Q. Song, J. Zhou, R. Feng, T. Chen, W. Ge, D.Z. Chen, S. Kevin Zhou, W. Wang, and J. Wu, Dual-level selective transfer learning for intrahepatic cholangiocarcinoma segmentation in non-enhanced abdominal CT.
文章选读：
- 2020.04.13: 刘蓬博 (Segmentation)
Paper-1: S. Wu, G. Wang, P. Tang, F. Chen, and L. Shi, “Convolution with even-sized kernels and symmetric padding,” in Advances in Neural Information Processing Systems, pp. 1194–1205, 2019.
Paper-2: D. Nie and D. Shen, “Adversarial confidence learning for medical image segmentation and synthesis,” International Journal of Computer Vision, pp. 1–20, 2020.
Paper-3: L. Yanwei, S. Lin, C. Yukang, L. Zeming, Z. Xiangyu, W. Xingang, S. Jian, et al., “Learning dynamic routing for semantic segmentation,” 2020.
Paper-3 :
- 2020.04.20: 李涵 (IOU)
Paper 1: R., Hamid, et al: Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression
Paper 2: Z. Zheng, et al: Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression
- 2020.04.27: 顾峰 (Diffeomorphic registration)
Paper 1: B. Avants B, et al: Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain, MIA 2018.
Paper 2: A. V. Dalca, et al: Unsupervised learning for fast probabilistic diffeomorphic registration, MICCAI 2018
Paper-3: W. Mok, et al: Fast Symmetric Diffeomorphic Image Registration with Convolutional Neural Networks, CVPR 2020.
- 2020.05.25: 高若尘 (GNN in Medical Image)
Paper 1: Z., Zhai, et al. : Linking Convolutional Neural Networks with Graph Convolutional Networks: Application in Pulmonary Artery-Vein Separation, GLMI,2019.
Paper 2：H. Yang, et al: Interpretable multimodality embedding of cerebral cortex using attention graph network for identifying bipolar disorder, MICCAI,2019.
Paper 3：X. Wang et al: Interactive 3D Segmentation Editing and Refinement via Gated Graph Neural Networks, GLMI, 2019.
- 2020.06.08:朱玖闻(AI and Informatics)
Paper 1: A. S. Panayides, et al: AI and Medical Imaging Informatics: Current Challenges and Future Directions. JBHI, 2020.
Paper 2: A. Chartsias, et al: Adversarial Image Synthesis for Unpaired Multi-Modal Cardiac Data. MICCAI, 2017.
Paper 3: L. Liu, et al: Multi-Task Deep Model With Margin Ranking Loss for Lung Nodule Analysis. IEEE TMI, 2020.
- 2020.06.15: 孙孟柯 (Asynchronous and fair in FL)
Paper-1: C. Xie, et al: Asynchronous Federated Optimization.
Paper-2: T. Li, et al: Fair Resource Allocation in Federated Learning.ICLR 2020.
- 2020.06.29: 林杨 (Detection)
Paper-1: T. Lin., et al: Focal Loss for Dense Object Detection. IEEE ICCV,2017.
Paper-2: Z. Tian., et al: FCOS: Fully Convolutional One-Stage Object Detection. IEEE CVPR,2019.
Paper-3: E. Xie., et al:PolarMask: Single Shot Instance Segmentation with Polar Representation. IEEE CVPR,2020.
谢谢关注MIRACLE 奇迹研究组！欢迎垂询zhoushaohua[at]ict.ac.cn。
MIRACLE长期寻找：
- 合作伙伴（医院、医疗设备商、科研机构等）；
- 助理研究员、博士后和工程人员等；
- 博士硕士研究生、实习生（全职或远程）等。