arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.04167cs.CVcs.AI
收藏

三维辐射组表示的不平衡感知自监督学习

Imbalance-Aware Self-Supervised Learning for 3D Radiomic Representations

Technical University of Munich(慕尼黑工业大学);University of Zurich(苏黎世大学);Southern University of Science and Technology(南方科技大学);ETH Zurich(苏黎世联邦理工学院);Macau University of Science and Technology(澳门科技大学);Klinikum rechts der Isar(伊萨尔河右岸医院)

Hongwei Li, Fei-Fei Xue, Krishna Chaitanya, Shengda Luo, Ivan Ezhov, Benedikt Wiestler, Jianguo Zhang, Bjoern Menze

摘要:

放射组表示可以量化医学图像数据中的感兴趣区域的属性。传统上,它们解释了形状、纹理和其他低级图像特征的预定义统计。或者,基于深度学习的表示源自监督学习,但需要来自专家的昂贵注释,并且经常遭受过拟合和数据不平衡问题。在这项工作中,我们解决了学习表示的3D医学图像的数据不平衡下的有效量化的挑战。我们提出了一个\emph{self-supervised}表示学习框架,学习高层次的功能,现有的放射组学功能的补充。具体来说,我们演示了如何使用3D Siamese网络以自监督方式学习图像表示。更重要的是,我们通过两种无监督策略来处理数据不平衡:a)样本重新加权,b)平衡训练批次的组成。当将我们学习的自监督特征与传统放射组学相结合时,我们在涵盖MRI和CT成像模式的脑肿瘤分类和肺癌分期任务中显示出显着改善。

英文摘要:

Radiomic representations can quantify properties of regions of interest in medical image data. Classically, they account for pre-defined statistics of shape, texture, and other low-level image features. Alternatively, deep learning-based representations are derived from supervised learning but require expensive annotations from experts and often suffer from overfitting and data imbalance issues. In this work, we address the challenge of learning representations of 3D medical images for an effective quantification under data imbalance. We propose a \emph{self-supervised} representation learning framework to learn high-level features of 3D volumes as a complement to existing radiomics features. Specifically, we demonstrate how to learn image representations in a self-supervised fashion using a 3D Siamese network. More importantly, we deal with data imbalance by exploiting two unsupervised strategies: a) sample re-weighting, and b) balancing the composition of training batches. When combining our learned self-supervised feature with traditional radiomics, we show significant improvement in brain tumor classification and lung cancer staging tasks covering MRI and CT imaging modalities.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑