CVPR每年都有不少医学影像相关文章，CVPR2021共有1663篇论文被接收，接收率为23.7%，其中有不少医学影像的文章，一起来看看吧。
参考：https://github.com/extreme-assistant/CVPR2021-Paper-Code-Interpretation
本系列将对部分文章进行深入解读，敬请关注！
[1] Automatic Vertebra Localization and Identification in CT by Spine Rectification and Anatomically-constrained Optimization
通过脊柱矫正和解剖学约束优化在CT中自动进行椎骨定位和识别
paper
[2] Deep Lesion Tracker: Monitoring Lesions in 4D Longitudinal Imaging Studies
深部病变追踪器：在4D纵向成像研究中监控病变
paper
[3] 3D Graph Anatomy Geometry-Integrated Network for Pancreatic Mass Segmentation, Diagnosis, and Quantitative Patient Management
用于胰腺肿块分割，诊断和定量患者管理的3D图形解剖学几何集成网络
paper
The pancreatic disease taxonomy includes ten types of masses (tumors or cysts). Previous work focuses on developing segmentation or classification methods only for certain mass types. Differential diagnosis of all mass types is clinically highly desirable but has not been investigated using an automated image understanding approach. We exploit the feasibility to distinguish pancreatic ductal adenocarcinoma (PDAC) from the nine other nonPDAC masses using multi-phase CT imaging. Both image appearance and the 3D organ-mass geometry relationship are critical. We propose a holistic segmentation-mesh-classification network (SMCN) to provide patient-level diagnosis, by fully utilizing the geometry and location information, which is accomplished by combining the anatomical structure and the semantic detection-by-segmentation network. SMCN learns the pancreas and mass segmentation task and builds an anatomical correspondence-aware organ mesh model by progressively deforming a pancreas prototype on the raw segmentation mask (i.e., mask-to-mesh). A new graph-based residual convolutional network (Graph-ResNet), whose nodes fuse the information of the mesh model and feature vectors extracted from the segmentation network, is developed to produce the patient-level differential classification results. Extensive experiments on 661 patients’ CT scans (five phases per patient) show that SMCN can improve the mass segmentation and detection accuracy compared to the strong baseline method nnUNet (e.g., for nonPDAC, Dice: 0.611 vs. 0.478; detection rate: 89% vs. 70%), achieve similar sensitivity and specificity in differentiating PDAC and nonPDAC as expert radiologists (i.e., 94% and 90%), and obtain results comparable to a multimodality test that combines clinical, imaging, and molecular testing for clinical management of patients.
胰腺疾病分类包括十种肿块（肿瘤或囊肿）。先前工作仅针对某些类型开发分割或分类方法。临床上非常需要对所有肿块类型进行鉴别诊断，但尚未使用自动图像理解方法进行鉴别诊断。我们利用多相CT成像技术来区分胰腺导管腺癌（PDAC）与其他九个非PDAC肿块。图像外观和3D器官-肿块几何关系都至关重要。我们提出了一种整体分割网分类网络（SMCN），通过充分利用几何结构和位置信息来提供患者级别的诊断，这是通过将解剖结构和语义 detection-by-segmentation 网络相结合来实现的。 SMCN学习胰腺和肿块的分割任务，并通过在原始分割 mask（即“mask-to-mesh”）上逐步变形胰腺原型，建立解剖上对应的器官网格模型。开发了一种新的基于图的残差卷积网络（Graph-ResNet），其节点融合了网格模型的信息和从分割网络中提取的特征向量，从而产生了患者级别的分类结果。在661位患者的CT扫描上进行的广泛实验（每位患者五个阶段）显示，与nnUNet相比，SMCN可以改善质量分割和检测准确性（例如，对于非PDAC，Dice：0.611 vs. 0.478；检测率：89％ vs. 70％），在区分PDAC和nonPDAC方面达到了与放射线专家相似的敏感性和特异性（94％和90％），并获得了与临床，影像学和分子检测相结合的多模式测试相当的结果。
[4] Multi-institutional Collaborations for Improving Deep Learning-based Magnetic Resonance Image Reconstruction Using Federated Learning
多机构协作改进基于深度学习的联合学习磁共振图像重建
paper | code
[5] DeepTag: An Unsupervised Deep Learning Method for Motion Tracking on Cardiac Tagging Magnetic Resonance Images
一种心脏标记磁共振图像运动跟踪的无监督深度学习方法
paper
[6] Discovering Hidden Physics Behind Transport Dynamics
在运输动力学背后发现隐藏物理
paper
[7] Multiple Instance Captioning: Learning Representations from Histopathology Textbooks and Articles
多实例字幕：从组织病理学教科书和文章中学习表示形式
paper
[8] FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space
在连续频率空间中通过情景学习进行医学图像分割的联合域泛化
paper | code
[9] XProtoNet: Diagnosis in Chest Radiography with Global and Local Explanations
使用全局和局部解释诊断胸部X光片
paper
[10] Brain Image Synthesis with Unsupervised Multivariate Canonical CSCℓ4Net
无监督多元规范CSCℓ4Net的脑图像合成
paper
[11] Confluent Vessel Trees with Accurate Bifurcations
分叉的融合容器树
paper
[12] DiNTS: Differentiable Neural Network Topology Search for 3D Medical Image Segmentation
DiNTS：用于3D医学图像分割的可区分神经网络拓扑搜索
paper
[13] Every Annotation Counts: Multi-label Deep Supervision for Medical Image Segmentation
每种注释都至关重要：医学图像分割的多标签深度监管
paper
