Fueled by recent advances in machine learning, there has been tremendous progress in the field of semantic segmentation for the medical image computing community. However, developed algorithms are often optimized and validated by hand based on one task only. In combination with small datasets, interpreting the generalizability of the results is often difficult. The Medical Segmentation Decathlon challenge addresses this problem, and aims to facilitate development of generalizable 3D semantic segmentation algorithms that require no manual parametrization. Such an algorithm was developed and is presented in this paper. It consists of a 3D convolutional neural network with encoder-decoder architecture employing residual-connections, skip-connections and multi-level generation of predictions. It works on anisotropic voxel-geometries and has anisotropic depth, i.e., the number of downsampling steps is a task-specific parameter. These depths are automatically inferred for each task prior to training. By combining this flexible architecture with on-the-fly data augmentation and little-to-no pre-- or postprocessing, promising results could be achieved. The code developed for this challenge will be available online after the final deadline at: https://github.com/ORippler/MSD_2018
MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis
Shanghai Jiao Tong University(上海交通大学)
我们提出了MedMNIST,一个10个预处理的医学开放数据集的集合。MedMNIST是标准化的,可对轻量级28 x 28图像执行分类任务,不需要背景知识。它涵盖了医学图像分析中的主要数据模式,在数据规模(从100到100,000)和任务(二进制/多类,有序回归和多标签)方面具有多样性。MedMNIST可用于医学图像分析中的教育目的、快速原型、多模态机器学习或AutoML。此外,MedMNIST Classification Decathlon旨在对所有10个数据集的AutoML算法进行基准测试;我们比较了几种基线方法,包括开源或商业AutoML工具。MedMNIST的数据集、评价代码和基线方法可在https://medmnist.github.io/上公开获得。
Auto Machine Learning for Medical Image Analysis by Unifying the Search on Data Augmentation and Neural Architecture
Sichuan University(四川大学)
自动数据增强是近年来研究的热点,其目的是自动制定工程增强策略。许多以前的自动增强方法利用密度匹配策略,通过评估测试时增强性能方面的政策。在本文中,我们从理论上和经验证明之间的不一致性的训练和验证集的小规模医学图像数据集,称为域内采样偏差。接下来,我们证明了域内采样偏差可能导致密度匹配的效率低下。针对这一问题,提出了一种改进的增广搜索策略--增广密度匹配(Augmented Density Matching),该策略从先验分布中随机抽取策略进行训练。此外,通过统一数据增强和神经结构的搜索,提出了一种高效的自动机器学习(AutoML)算法。实验结果表明,所提出的方法优于最先进的方法MedMNIST,一个开创性的基准设计的AutoML在医学图像分析。
King’s College London(伦敦国王学院);German Cancer Research Center (DKFZ)(德国癌症研究中心);University of Heidelberg(海德堡大学);University of Pennsylvania(宾夕法尼亚大学);National Cancer Institute(美国国家癌症研究所);Vanderbilt University(范德堡大学);Radboud University Medical Center(拉德堡德大学医学中心);University of Zurich(苏黎世大学);DeepMind(深度思维)
Universal and Extensible Language-Vision Models for Organ Segmentation and Tumor Detection from Abdominal Computed Tomography
City University of Hong Kong(香港城市大学);Johns Hopkins University(约翰斯·霍普金斯大学);University of California, San Francisco(加州大学旧金山分校);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Chinese University of Hong Kong(香港中文大学);NVIDIA(英伟达)
Guiding the Guidance: A Comparative Analysis of User Guidance Signals for Interactive Segmentation of Volumetric Images
Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院);HIDSS4Health - Helmholtz Information and Data Science School for Health(HIDSS4Health——亥姆霍兹健康信息与数据科学学院);Institute for AI in Medicine, University Hospital Essen(埃森大学医院医学人工智能研究所);Cancer Research Center Cologne Essen (CCCE), University Medicine Essen(埃森大学医学院科隆埃森癌症研究中心(CCCE))
MSAD-Net: Multiscale and Spatial Attention-based Dense Network for Lung Cancer Classification
Computer Science and Engineering Department, NIIT University, Jaipur, India(尼特大学计算机科学与工程系,印度斋普尔);CSE department, Christ (Deemed to be University), Bangalore, India(基督大学(认定为大学)计算机科学与工程系,印度班加罗尔);Faculty of Computing, Engineering and the Built Environment, Birmingham City University, UK(伯明翰城市大学计算、工程与建筑环境学院,英国)
Mirror U-Net: Marrying Multimodal Fission with Multi-task Learning for Semantic Segmentation in Medical Imaging
Institute for Anthropomatics and Robotics (IAR), Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院人机交互与机器人研究所);German Cancer Consortium (DKTK)(德国癌症联盟);HIDSS4Health - Helmholtz Information and Data Science School for Health(亥姆霍兹健康信息与数据科学学院(HIDSS4Health));Institute for AI in Medicine (IKIM), University Hospital Essen(埃森大学医院医学人工智能研究所(IKIM));University Hospital Essen(埃森大学医院);West German Cancer Center, Cancer Research Center Cologne Essen (CCCE), University Medicine Essen(埃森大学医学院西德癌症中心、科隆埃森癌症研究中心(CCCE))
Frequency Disentangled Learning for Segmentation of Midbrain Structures from Quantitative Susceptibility Mapping Data
Sorbonne Université(索邦大学);Institut du Cerveau - Paris Brain Institute - ICM(巴黎脑研究所);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);Inserm(法国国家健康与医学研究院);AP-HP(巴黎公立医院集团);Hôpital de la Pitié Salpêtrière(巴黎萨尔佩特里埃医院);Centre de NeuroImagerie de Recherche-CENIR(神经影像研究中心)
More than Encoder: Introducing Transformer Decoder to Upsample
South China University of Technology(华南理工大学);Guangdong Academy of Medical Sciences(广东省医学科学院);Sun Yat-sen University(中山大学);Beijing Institute of Technology(北京理工大学)