Stacked dilated convolutions and asymmetric architecture for U-Net-based medical image segmentation.
Stacked dilated convolutions and asymmetric architecture for U-Net-based medical image segmentation.
复制标题
DOI：
10.1016/j.compbiomed.2022.105891
复制
发表时间：
2022-09
影响因子：
7.7
通讯作者：
中科院分区：
文献类型：
--
作者：
深度学习已被广泛用于医学图像分割。最常用的U-Net及其变体通常具有两个共同特征，但缺乏有效性的确凿证据。首先，每个块（即，相同分辨率的特征图的连续卷积）输出来自最后卷积的特征图，限制了感受野的多样性。第二，网络具有对称结构，其中编码器和解码器路径具有相似数量的信道。我们探索了两种新的改进：一种是堆叠扩张操作，从多尺度感受野输出特征图，以取代连续卷积;一种是解码器路径中通道较少的非对称结构。我们开发了两种新的模型：使用堆叠扩张操作的U-Net（SDU-Net）和非对称SDU-Net（ASDU-Net）。我们使用公开和私人数据集来评估所提出的模型的有效性。广泛的实验证实，SDU-Net在使用更少参数（U-Net的40%）的情况下优于或达到了与最先进技术相似的性能。ASDU-Net进一步将模型参数降低到U-Net的20%，性能与SDU-Net相当。总之，堆叠膨胀操作和非对称结构是有希望提高性能的U-网及其变种。
Deep learning has been widely utilized for medical image segmentation. The most commonly used U-Net and its variants often share two common characteristics but lack solid evidence for the effectiveness. First, each block (i.e., consecutive convolutions of feature maps of the same resolution) outputs feature maps from the last convolution, limiting the variety of the receptive fields. Second, the network has a symmetric structure where the encoder and the decoder paths have similar numbers of channels. We explored two novel revisions: a stacked dilated operation that outputs feature maps from multi-scale receptive fields to replace the consecutive convolutions; an asymmetric architecture with fewer channels in the decoder path. Two novel models were developed: U-Net using the stacked dilated operation (SDU-Net) and asymmetric SDU-Net (ASDU-Net). We used both publicly available and private datasets to assess the efficacy of the proposed models. Extensive experiments confirmed SDU-Net outperformed or achieved performance similar to the state-of-the-art while using fewer parameters (40% of U-Net). ASDU-Net further reduced the model parameters to 20% of U-Net with performance comparable to SDU-Net. In conclusion, the stacked dilated operation and the asymmetric structure are promising for improving the performance of U-Net and its variants.
登录
查看更多内容
DOI：
10.1007/978-3-319-67558-9_28
发表时间：
2017-09-09
期刊：
Deep learning in medical image analysis and multimodal learning for clinical decision support : Third International Workshop, DLMIA 2017, and 7th International Workshop, ML-CDS 2017, held in conjunction with MICCAI 2017 Quebec City, QC,..
影响因子：
--
作者：
Sudre CH;Li W;Vercauteren T;Ourselin S;Jorge Cardoso M
通讯作者：
Jorge Cardoso M
影响因子：
10.9
作者：
Abu Anas, Emran Mohammad;Mousavi, Parvin;Abolmaesumi, Purang
通讯作者：
Abolmaesumi, Purang
影响因子：
3.4
作者：
Devalla, Sripad Krishna;Renukanand, Prajwal K.;Girard, Michael J. A.
通讯作者：
Girard, Michael J. A.
影响因子：
3.3
作者：
Singh, Raman Preet;Gupta, Savita;Acharya, U. Rajendra
通讯作者：
Acharya, U. Rajendra
影响因子：
6.1
作者：
Al-Masni, Mohammed A.;Al-antari, Mugahed A.;Kim, Tae-Seong
通讯作者：
Kim, Tae-Seong