• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Deep learning in medical image analysis and multimodal learning for clinical decision support >Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations
【24h】

Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations

机译：广义骰子重叠作为高度不平衡细分的深度学习损失功能

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Deep-learning has proved in recent years to be a powerful tool for image analysis and is now widely used to segment both 2D and 3D medical images. Deep-learning segmentation frameworks rely not only on the choice of network architecture but also on the choice of loss function. When the segmentation process targets rare observations, a severe class imbalance is likely to occur between candidate labels, thus resulting in sub-optimal performance. In order to mitigate this issue, strategies such as the weighted cross-entropy function, the sensitivity function or the Dice loss function, have been proposed. In this work, we investigate the behavior of these loss functions and their sensitivity to learning rate tuning in the presence of different rates of label imbalance across 2D and 3D segmentation tasks. We also propose to use the class re-balancing properties of the Generalized Dice overlap, a known metric for segmentation assessment, as a robust and accurate deep-learning loss function for unbalanced tasks.
机译：近年来，深度学习已被证明是用于图像分析的强大工具，如今已被广泛用于分割2D和3D医学图像。深度学习细分框架不仅依赖于网络体系结构的选择，还依赖于损失函数的选择。当分割过程针对稀少的观察结果时，候选标签之间可能会出现严重的类不平衡，从而导致性能欠佳。为了减轻这个问题，已经提出了诸如加权交叉熵函数，灵敏度函数或骰子损失函数之类的策略。在这项工作中，我们研究了在2D和3D分割任务中标签失衡率不同的情况下，这些损失函数的行为及其对学习率调整的敏感性。我们还建议使用广义骰子重叠的类重新平衡属性（一种用于细分评估的已知度量），作为不平衡任务的健壮且准确的深度学习损失函数。

著录项

  • 来源
  • 会议地点 Quebec City(CA)
  • 作者

    Carole H. Sudre; Wenqi Li; Tom Vercauteren; Sebastien Ourselin; M. Jorge Cardoso;

  • 作者单位

    Translational Imaging Group, CMIC, University College London, London NW1 2HE, UK,Dementia Research Centre, UCL Institute of Neurology, London WC1N 3BG, UK;

    Translational Imaging Group, CMIC, University College London, London NW1 2HE, UK;

    Translational Imaging Group, CMIC, University College London, London NW1 2HE, UK;

    Translational Imaging Group, CMIC, University College London, London NW1 2HE, UK,Dementia Research Centre, UCL Institute of Neurology, London WC1N 3BG, UK;

    Translational Imaging Group, CMIC, University College London, London NW1 2HE, UK,Dementia Research Centre, UCL Institute of Neurology, London WC1N 3BG, UK;

  • 会议组织
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号