基于Dense-UNet++的关节滑膜磁共振图像分割
王振宇,王颖珊,毛瑾玲,马伟伟,路青,石洁,汪红志

Magnetic Resonance Images Segmentation of Synovium Based on Dense-UNet++
Zhen-yu WANG,Ying-shan WANG,Jin-ling MAO,Wei-wei MA,Qing LU,Jie SHI,Hong-zhi WANG
表3 不同分割模型的验证集与训练集DSC曲线的振荡系数
Table 3 Oscillation coefficients of DSC curves of different models
网络结构 激活函数 是否使用密集模块 DSC曲线振荡系数
UNet ReLu 2.534×10-4
UNet Swish 1.439×10-4
ResUNet ReLu 3.295×10-4
VGG-UNet++ ReLu 2.321×10-4
Dense-UNet++ ReLu 9.412×10-5
Dense-UNet++(本文方法) Swish 2.835×10-5