老中医5555-优快云博客

原创修改nnUNet的部分组件（参考作者提供的doc）

修改nnUNet的部分组件（参考作者提供的doc）下文主要是作者在nnUNet.documentation.extending_nnunet的翻译https://github.com/MIC-DKFZ/nnUNet/tree/master/documentation/extending_nnunet.md首先确保安装nnunet的方法为从github拖最新代码到本地，并使用pip install -e .进行安装，这样可确保本地有一份源码的拷贝根据参数不同（推断，蓝图，经验参数/inferred,

2020-12-01 22:34:09 2863 7

原创医学分割论文:Segmentation of CT thoracic organs by multi-resolution VB-nets

医学分割论文:Segmentation of CT thoracic organs by multi-resolution VB-netsAuther: Miaofei Han. Institution:1Shanghai United Imaging Intelligence Inc. Publish year: 2019论文地址：SegTHOR2019在本篇论文中，作者提出了VB-Net结构，采用了多尺度分割网络，参与了SegTHOR胸部器官分割挑战赛，并在四个危及器官勾画中取得第一名Overv

2020-12-01 22:33:50 965

原创医学图像分割论文:Unified Representation Learning for Efficient Medical Image Analysis

医学图像分割论文:Unified Representation Learning for Efficient Medical Image AnalysisAuther: Ghada Zamzmi, Institution: National Institutes of Health, Publish year: 2020论文地址：https://arxiv.org/abs/2006.11223Key point:医学图像分析通常包括图像增强、检测、分割和分类等任务，通常通过单独的机器学习方法来实现

2020-09-22 08:47:45 1108

原创图像分割论文:Efficient, high-performance pancreatic segmentation using multi-scale feature extraction

图像分割论文:Efficient, high-performance pancreatic segmentation using multi-scale feature extractionAuther: Moritz Knolle, Institution:Technical University of Munich, Publish year: 2020论文地址：https://arxiv.org/abs/:2009.00872代码地址：https://github.com/TUM-AIMED/M

2020-09-22 08:47:32 280

原创图像分割论文:Quality-aware semi-supervised learning for CMR segmentation

图像分割论文:Quality-aware semi-supervised learning for CMR segmentationAuther: Bram Ruijsink, Institution:King’s College London, Publish year: 2020论文地址：https://arxiv.org/abs/2009.00584Key point:医学图像分割目前存在的问题是难以获取优质的带标注数据，而数据增强和半监督式的方法有效性有限。作者基于这样一个前提：医学分割

2020-09-22 08:47:23 275

原创图像分割论文:ARPM-net

图像分割论文:ARPM-net: A novel CNN-based adversarial method with Markov Random Field enhancement for prostate and organs at risk segmentation in pelvic CT imagesAuther: Zhuangzhuang Zhang, Institution:Washington University, Publish year: 2020论文地址：https://arxi

2020-09-22 08:47:15 271 1

原创 Accurate 2D soft segmentation of medical image via SoftGAN network

图像分割论文:Accurate 2D soft segmentation of medical image via SoftGAN networkAuther: Changwei Wang, Institution:Chinese Academy of Sciences, Publish year: 2020论文地址：https://arxiv.org/abs/2007.14556v1Key point:针对二维肺结节分割和CT超分辨任务，提出了CasGAN，首次引入了精确描述病变的语义软分割形式

2020-09-22 08:46:59 224

原创图像分割论文(nnUNet):Automated Design of Deep Learning Methods for Biomedical Image Segmentation

图像分割论文:Automated Design of Deep Learning Methods for Biomedical Image SegmentationAuther: Fabian Isensee, Institution:German Cancer Research Center, Publish year: 2020论文地址：https://arxiv.org/abs/1904.08128v2github: https://github.com/MIC-DKFZ/nnUNetKey

2020-09-22 08:46:51 2992

原创医学图像分割论文:FocusNet: Imbalanced Large and Small Organ Segmentation with an e2e DNN for Head CT

医学图像分割论文:FocusNet: Imbalanced Large and Small Organ Segmentation with an End-to-End Deep Neural Network for Head and Neck CT ImagesAuther: Yunhe Gao, Institution: SenseTime Research, Publish year: 2019论文地址：https://arxiv.org/abs/1907.12056v1Key point:提出

2020-09-22 08:46:42 797

原创医学图像分类论文:Deep Learning for Identifying Metastatic Breast Cancer

医学图像分类论文:Deep Learning for Identifying Metastatic Breast CancerAuther: Dayong Wang, Institution: Harvard Medical School, Publish year: 2016论文地址：https://arxiv.org/abs/1606.05718v1Key point:ISBI challenge: 淋巴结活检图像中转移性乳腺癌的自动检测，体现了深度学习在自动诊断中的应用将原始图像划分为小的p

2020-09-22 08:46:34 1386

原创医学分割论文:Vox2Vox: 3D-GAN for Brain Tumour Segmentation

医学分割论文:Vox2Vox: 3D-GAN for Brain Tumour SegmentationAuther: Marco Domenico, Institution:Department of Biomedical Engineering, Publish year: 2020论文地址：https://arxiv.org/abs/2003.13653v1key point使用3D volume-to-volume GAN用于脑肿瘤分割。当生成器loss是鉴别器loss5倍时获得最佳分割效果

2020-09-22 08:46:25 901

原创医学分割论文:Deep Learning Based Brain Tumor Segmentation: a survey

医学分割论文:Deep Learning Based Brain Tumor Segmentation: a surveyAuther: Zhihua Liu, Institution:University of Leicester, Publish year: 2020论文地址：https://arxiv.org/abs/2007.09479v2综述类论文背景脑部肿瘤难以分割的原因：脑肿瘤可能出现在脑组织中的任何位置，其形状和大小均不确定；另一方面，脑肿瘤和正常组织之间存在相互浸润的边界，使得边

2020-09-22 08:46:11 1052

原创 nnUNet安装踩坑记录

nnUNet安装踩坑记录论文地址：https://arxiv.org/abs/1809.10486https://github.com/MIC-DKFZ/nnUNetnnUNet采用2D，3D和级联UNet网络结构，最大的亮点在于可以自适应调参，可以适用于多种数据集的图像分割训练，尤其是医学图像分割训练，因此想尝试部署该框架，对CT和MR的癌症靶区进行分割部署环境操作系统：windows 10GPU：nvidia 1080 8Gpython 环境：使用anaconda建立独立虚拟环境，pyt

2020-07-31 21:13:22 3011

原创 nnUNet安装踩坑-自适应医学图像分割框架

nnunet是由德国学者提出来的一个用于医学图像分割的通用自适应框架项目地址：https://github.com/MIC-DKFZ/nnUNet最近想在电脑上部署一下，顺便踩踩坑，感觉其中最大的坑就是apex的安装，深度学习环境的部署可以看上一篇文章硬件环境HP Z8G4工作站，RTX2080ti显卡Ubuntu 1804.4 桌面版本Nvidia graphic driver 450.57安装CUDA和CuDNN这部分还算是比较简单，从官网下载CUDA10.0和对应的CuDNN 7.

2020-07-31 21:13:15 5092 1

原创 nnUNet使用指南＜1＞:使用自己的数据集做分割训练

首先，看了看github上的readme，确实挺长。目前数据都有，就按照步骤一步一步来吧设置路径nnUNet需要设置三个路径：nnUNet_raw_data_base: 用于存放原始数据和crop后的数据。该目录至少包含一个nnUNet_raw_data子文件夹nnUNet_preprocessed：用于存放预处理数据，该数据将来会用于训练，建议放在SSD下RESULTS_FOLDER：用于存放训练模型使用文本编辑器打开.bashrc文件，将以上路径加入环境变量sudo vim ~/.b

2020-07-31 21:13:05 7312 36

原创深度学习环境配置（ubuntu 1804）真.从零开始

深度学习环境配置（ubuntu 1804）真.从零开始最近接到个任务，需要在一台HP Z8G4工作站上配置深度学习系统，正好来记录一下安装过程和安装过程中遇到的坑。首先，先理一下安装思路：系统采用ubuntu1804版本，至少这个版本兼容性比较好，之前的1604版本，有时会遇到一些问题。之后再完成硬盘挂载（一共三块硬盘），驱动安装和相关环境配置（目前是使用anaconda）因为不需要双系统，硬盘本身没有系统，因此这也算是从零开始了安装系统本来是打算直接装server版本，但是我这边上网需要浏览器访问网

2020-07-30 22:09:58 654

原创医学图像分割论文:Ischemic Stroke Lesion Seg in CT Perfusion Scans using Pyramid Pooling and Focal Loss

医学图像分割论文:Ischemic Stroke Lesion Segmentation in CT Perfusion Scans using Pyramid Pooling and Focal LossAuther: Mazdak Abulnaga, Institution: MIT & Philips, Publish year: 2018论文地址：https://arxiv.org/abs/1811.01085v1Key point:在PSPNet（采用pyramid pooling

2020-07-30 22:09:19 760 1

原创医学图像分割论文:MS-NAS: Multi-Scale Neural Architecture Search for Medical Image Segmentation

医学图像分割论文:MS-NAS: Multi-Scale Neural Architecture Search for Medical Image SegmentationAuther: Xingang Yan, Institution:Zhejiang University, Publish year: 2020论文地址：https://arxiv.org/abs/2007.06151Key point:将网络结构搜索（NeuralArchitecture Search, NAS）应用于医学图像分

2020-07-30 22:09:10 1242

原创医学图像分割论文:Two-Stage Cascaded U-Net: 1st Place Solution to BraTS Challenge 2019 Segmentation Task

医学图像分割论文:Two-Stage Cascaded U-Net: 1st Place Solution to BraTS Challenge 2019 Segmentation TaskAuther: Zeyu Jiang, Institution:South China University of Technology, Publish year: 2020论文地址：https://sci-hub.tw/10.1007/978-3-030-46640-4_22Key point:提出了end-

2020-07-30 22:08:58 2560 7

原创医学分割论文:UNet++: A Nested U-Net Architecture for Medical Image Segmentation

医学分割论文:UNet++: A Nested U-Net Architecture for Medical Image SegmentationAuther: Zongwei Zhou, Institution:Arizona State University, Publish year: 2018论文地址：https://arxiv.org/abs/1807.10165https://github.com/MrGiovanni/UNetPlusPlus目前存在的问题：传统的UNet中的ski

2020-07-30 22:08:48 426

原创医学分割论文:nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation

医学分割论文:nnU-Net: Self-adapting Framework for U-Net-Based Medical Image SegmentationAuther: Fabian Isensee, Institution:German Cancer Research Center, Publish year: 2018论文地址：https://arxiv.org/abs/1809.10486https://github.com/MIC-DKFZ/nnUNet目前存在的问题：UNet

2020-07-30 22:08:28 1016

原创图像分割论文:Automatic liver tumor segmentation in CT with FCNN and object-based postprocessing

图像分割论文:Automatic liver tumor segmentation in CT with fully convolutional neural networks and object-based postprocessingAuther: Grzegorz Chlebus, Institution:Fraunhofer Institute for Medical Image Computing MEVIS, Publish year: 2018论文地址：DOI:10.1038/s4159

2020-07-30 22:08:18 440

原创医学分割论文:Deep Learning for Automated Contouring of Primary Tumor Volumes by MRI for NPC

医学分割论文:Deep Learning for Automated Contouring of Primary Tumor Volumes by MRI for Nasopharyngeal CarcinomaAuther: Li Lin, Institution:Sun Yat-sen University Cancer Center, Publish year: 2019论文地址：https://doi.org/10.1148/radiol.2019182012https://github.co

2020-07-30 22:08:08 408

原创医学分割论文:Automated Segmentation of Colorectal Tumor in 3D MRI Using 3D MSD CNN network

医学分割论文:Automated Segmentation of Colorectal Tumor in 3D MRI Using 3D Multiscale Densely Connected Convolutional Neural NetworkAuther: Hussain Soomro, Institution:University of Roma Tre, Publish year: 2019论文地址：https://doi.org/10.1155/2019/1075434Key poin

2020-07-30 22:07:54 377

原创图像分割论文:Beyond CNNs: Exploiting Further Inherent Symmetries in Medical Images for Segmentation

图像分割论文:Beyond CNNs: Exploiting Further Inherent Symmetries in Medical Images for SegmentationAuther: Shuchao Pang, Institution:Macquarie University, Publish year: 2020论文地址：https://arxiv.org/abs/2005.03924Key point:提出了目前CNN-based方法的局限性：CNN只能利用平移不变性，而忽略了

2020-07-30 22:07:24 230

原创医学分割论文:Automatic Liver and Lesion Segmentation in CT Using Cascaded FCNN and 3D-CRF

医学分割论文:Automatic Liver and Lesion Segmentation in CT Using Cascaded Fully Convolutional Neural Networks and 3D Conditional Random FieldsAuther: Ferdinand Christ, Institution:Technische Universitat Munchen, Publish year: 2016论文地址：https://arxiv.org/abs/16

2020-06-19 20:09:07 608

原创医学分割论文:Recurrent Residual Convolutional Neural Network based on U-Net for Medical Image Segmentation

医学分割论文:Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image SegmentationAuther: Md Alom. Institution:University of Dayton. Publish year: 2019论文地址：https://arxiv.org/abs/1802.06955在本篇论文中，作者提出了U-Net的改进模型：循环卷积网络模型（Recur

2020-06-19 20:08:59 970

原创医学分割论文:Multi-scale guided attention for medical image segmentation

医学分割论文:Multi-scale guided attention for medical image segmentationAuther: Ashish Sinha, Institution:Indian Institute of Technology Roorkee, Publish year: 2019论文地址：https://arxiv.org/abs/1906.02849github: https://github.com/sinAshish/Multi-Scale-Attention

2020-06-19 20:08:46 990

原创图像分割论文:Boundary Loss for Remote Sensing Imagery semantic segmentation

图像分割论文:Boundary Loss for Remote Sensing Imagery semantic segmentationAuther: Alexey Bokhovkin, Institution:Skoltech, Moscow, Publish year: 2019论文地址：https://arxiv.org/abs/1905.07852v1目前存在的问题：对遥感地图进行分析时，对目标实例的分割，尤其是对目标边界的确定至关重要。但目前分割任务中广泛使用的loss functio

2020-06-19 20:08:33 5053

jlch5555的博客