Deep Learning for Carotid Plaque Segmentation using a Dilated U-Net Architecture.
Deep Learning for Carotid Plaque Segmentation using a Dilated U-Net Architecture.
复制标题
DOI：
10.1177/0161734620951216
复制
发表时间：
2020-07
影响因子：
2.3
通讯作者：
Varghese T
中科院分区：
文献类型：
--
作者：
Meshram NH;Mitchell CC;Wilbrand S;Dempsey RJ;Varghese T
这项工作提出了使用深度学习在超声纵向 B 模式图像中进行颈动脉斑块分割。我们报告了 101 名严重狭窄颈动脉斑块患者。将标准 U-Net 与扩张 U-Net 架构进行比较，其中扩张卷积层用于瓶颈。实现了带有边界框的全自动和半自动方法。量化由于边界框错误而导致的斑块分割性能下降。我们发现，边界框显着提高了网络的性能，自动斑块分割的 U-Net Dice 系数为 0.48，半自动斑块分割的 U-Net Dice 系数为 0.83。与经验丰富的超声医师对相同斑块进行手动分割相比，扩张的 U-Net 也获得了类似的结果，自动 Dice 系数为 0.55，半自动为 0.84。两个维度中边界框的 5% 误差将 U-Net 和扩张 U-Net 的 Dice 系数分别降低至 0.79 和 0.80。
Carotid plaque segmentation in ultrasound longitudinal B-mode images using deep learning is presented in this work. We report on 101 severely stenotic carotid plaque patients. A standard U-Net is compared with a dilated U-Net architecture in which the dilated convolution layers were used in the bottleneck. Both a fully automatic and a semi-automatic approach with a bounding box was implemented. The performance degradation in plaque segmentation due to errors in the bounding box is quantified. We found that the bounding box significantly improved the performance of the networks with U-Net Dice coefficients of 0.48 for automatic and 0.83 for semi-automatic segmentation of plaque. Similar results were also obtained for the dilated U-Net with Dice coefficients of 0.55 for automatic and 0.84 for semi-automatic when compared to manual segmentations of the same plaque by an experienced sonographer. A five percent error in the bounding box in both dimensions reduced the Dice coefficient to 0.79 and 0.80 for U-Net and dilated U-Net respectively.
登录
查看更多内容
影响因子：
3.8
作者：
Orlando, Nathan;Gillies, Derek J.;Fenster, Aaron
通讯作者：
Fenster, Aaron
影响因子：
1.4
作者：
Meshram, N. H.;Mitchell, C. C.;Varghese, T.
通讯作者：
Varghese, T.
影响因子：
3.2
作者：
Loizou, C. P.;Pattichis, C. S.;Nicolaides, A.
通讯作者：
Nicolaides, A.
影响因子：
10.6
作者：
Leclerc, Sarah;Smistad, Erik;Bernard, Olivier
通讯作者：
Bernard, Olivier
DOI：
10.1007/s11548-020-02158-3
发表时间：
2020-04-29
影响因子：
3
作者：
Amiri, Mina;Brooks, Rupert;Rivaz, Hassan
通讯作者：
Rivaz, Hassan