• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Medical Image Computing and Computer-Assisted Intervention >ARS-Net: Adaptively Rectified Supervision Network for Automated 3D Ultrasound Image Segmentation
【24h】

ARS-Net: Adaptively Rectified Supervision Network for Automated 3D Ultrasound Image Segmentation

机译：ARS-Net：自动化3D超声图像分割的自适应纠正监控网络

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

摘要

3D Ultrasound (3DUS) has been widely used in clinical diagnosis. However, volume segmentation of 3DUS is very challenging due to relatively poor image quality and usually small datasets. We propose an efficient and robust method (ARS-Net) for single organ segmentation in 3DUS. Our contributions are twofold, (ⅰ) We propose a 2.5D framework based on a 2D segmentation network with 2.5D input which can provide more contextual and spatial information. The proposed framework also overcomes the limitation of 3D networks, such as small datasets, lack of pre-trained models and high memory cost, (ⅱ) To further enhance the performance in low signal-to-noise ratio (SNR) regions, we incorporate a new mechanism of adaptively rectified supervision (ARS) into the proposed 2.5D framework at training stage. Specifically, both pixel-wise reweighted dice loss and image-wise shape regu-larization loss are applied to improve the sensitivity and the specificity of segmentation. The experiment results on two representative and challenging datasets of 3DUS show that the proposed ARS-Net outperforms state-of-the-art methods with higher accuracy but lower complexity. The proposed novel network is robust to small datasets and can provide an accurate and fast volume segmentation tool for 3DUS.
机译：3D超声（3DUS）已广泛用于临床诊断。然而，由于图像质量较差和通常小的数据集，3DU的体积分割非常具有挑战性。我们为3DU中的单个器官分割提出了一种高效且稳健的方法（ARS-NET）。我们的贡献是双重的，（Ⅰ）我们提出了一个基于2D分段网络的2.5D框架，其中2.5D输入可以提供更多的上下文和空间信息。所提出的框架还克服了3D网络的限制，如小型数据集，缺乏预先训练的模型和高记忆成本，（Ⅱ），进一步提高了低信噪比（SNR）区域的性能，我们纳入了自适应纠正监督（ARS）进入培训阶段建议2.5D框架的新机制。具体地，施加像素 - 明智的重量丢失和图像明智形状的损失，以改善分割的灵敏度和特异性。实验结果对3DU的两个代表性和具有挑战性的数据集表明，所提出的ARS-net优于最先进的方法，具有更高的精度，但复杂性较低。所提出的新型网络对小型数据集具有鲁棒，可以为3DU提供准确和快速的分段工具。

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