• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Conference on Medical Imaging: Image Processing >Improving T_(1w) MRI-based brain tumor segmentation using cross-modal distillation
【24h】

Improving T_(1w) MRI-based brain tumor segmentation using cross-modal distillation

机译：使用跨型蒸馏改善T_（1W）MRI的脑肿瘤分割

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Although multi-modal imaging tends to improve the segmentation and classification performance in the field of medical image processing, lacking certain modalities at test time limits its clinical applicability. In this paper, we explored the ability of cross-modal distillation for increasing the performance of T_(1w) MRI-based brain tumor segmentation. More specifically, we considered having high resolution T_(1w) and T_(2w) MRI sequences available for training while having only a high resolution T_(1w) MRI sequence available at test time. We investigated the efficacy of the proposed method to improve the whole tumor segmentation using the BRATS 2018 dataset. Both cross-modal knowledge distillation and cross-modal feature distillation approaches were confirmed to enrich the representation of the T_(1w) MRI sequence by learning from the representation of the more informative T_(2w) MRI sequence during training, thereby improving the mean Dice scores by 6.14 % and 7.02 %, respectively.
机译：虽然多模态成像倾向于改善医学图像处理领域的分割和分类性能，但缺乏在测试时间下的某些模式限制其临床适用性。 在本文中，我们探讨了跨模态蒸馏来增加T_（1W）MRI的脑肿瘤细分性能的能力。 更具体地，我们考虑具有可用于训练的高分辨率T_（1W）和T_（2W）MRI序列，同时仅在测试时间可用的高分辨率T_（1W）MRI序列。 我们调查了使用Brats 2018 DataSet改善整个肿瘤细分的疗效。 确认跨模型知识蒸馏和跨模型特征蒸馏方法通过在训练期间从更具信息丰富的T_（2W）MRI序列的表示来丰富T_（1W）MRI序列的表示，从而改善平均骰子 分别得分6.14％和7.02％。

著录项

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