北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2023, Vol. 46 ›› Issue (2): 91-97.

• 模式识别与图像处理 • 上一篇    下一篇

基于RAPNet的脑肿瘤MRI图像三维分割

胡敏1,熊思1,黄宏程1,张光华2,王春芳3   

  1. 1. 重庆邮电大学
    2. 山西智能大数据产业技术创新研究院
    3. 山西医科大学第一医院
  • 收稿日期:2022-04-20 修回日期:2022-05-24 出版日期:2023-04-28 发布日期:2023-05-14
  • 通讯作者: 黄宏程 E-mail:huanghc@cqupt.edu.cn
  • 基金资助:
    国家自然科学基金;山西省回国留学人员科研资助项目;山西省重点研发项目;山西省卫健委研究基金项目

3D Segmentation of Brain Tumor MRI Image based on RAPNet

  • Received:2022-04-20 Revised:2022-05-24 Online:2023-04-28 Published:2023-05-14
  • Contact: Hong-Cheng Huang E-mail:huanghc@cqupt.edu.cn

摘要: 针对用于全自动脑肿瘤分割的传统深度卷积神经网络存在多尺度病变处理能力较弱问题,该文使用改进的3D递归残差卷积单元构建特征学习的主干,提高特征学习的空间相关性并缓解网络模型过于复杂造成的网络退化和梯度弥散;由具有不同膨胀率的3D空洞卷积和跨模型注意力机制构建分层特征金字塔,结合上下文特征以提高整体模型对不同大小肿瘤的识别能力;结合多层特征图对肿瘤进行辅助预测,以获得最终分割结果。在BraTS 2019数据集上进行的实验表明,该文方法在分割WT、TC、ET取得的平均DSC值分别为0.897、0.852、0.823,与现有高效的脑肿瘤分割方法相比在学习病变的多尺度特征方面具有更好的效果。

关键词: 脑肿瘤分割, 特征金字塔, 3D递归残差卷积单元, 注意力机制, 3D空洞卷积

Abstract: In view of the weak ability of multi-scale lesion processing ability of the traditional deep convolutional neural networks for fully automatic brain tumor segmentation, the improved recurrent residual convolutional units are used to build the backbone for feature learning to improve the spatial relevance of feature learning and alleviate the network degradation and gradient dispersion caused by too complex network model. The hierarchical feature pyramid is constructed by 3D atrous-convolution with different expansion rates and cross model attention mechanism, combined with context features to improve the recognition ability of the overall model for tumors of different sizes. Combined with multi-layer feature map, the tumor is predicted to obtain the final segmentation result. Abundant ablation experiments carried on BraTS 2019 datasets demonstrate that the average DSC values of WT, TC and ET were 0.897, 0.852 and 0.823 respectively. Compared with the existing efficient brain tumor segmentation methods, RAPNet has better effect in learning the multi-scale features of lesions.

Key words: brain tumor segmentation, feature pyramid, 3D recurrent residual convolution unit, attention mechanism, 3D atrous-convolution

中图分类号: