基于生成对抗网络的膝关节模型构建与局部比吸收率估计

任宏晋,马岩,肖亮

Knee Joint Model Construction and Local Specific Absorption Rate Estimation Based on Generative Adversarial Networks

REN Hongjin,MA Yan,XIAO Liang

表1 各种方法的分割结果评价指标（所提方法、U-Net、Attention U-Net相比人工标注结果）

Table 1 Evaluation indexes of the segmentation results of various methods (the proposed method, U-Net, Attention U-Net compared with the manual labeling results)

分割方法 评价指标 肌肉 脂肪 骨骼 所提方法 DCC 0.8798 0.9135 0.9022 TPR 0.8915 0.9198 0.9114 U-Net DCC 0.8472 0.9040 0.9006 TPR 0.8841 0.8785 0.8921 Attention U-Net DCC 0.8578 0.9022 0.9058 TPR 0.9059 0.8611 0.8991