Rethinking Pooling Operation for Liver and Liver-Tumor Segmentations

IF 1.3 Q3 ENGINEERING, ELECTRICAL & ELECTRONIC Frontiers in signal processing Pub Date : 2022-01-10 DOI:10.3389/frsip.2021.808050
Ju-hong Lei, Tao Lei, Weiqiang Zhao, Min-Qi Xue, Xiaogang Du, A. Nandi
{"title":"Rethinking Pooling Operation for Liver and Liver-Tumor Segmentations","authors":"Ju-hong Lei, Tao Lei, Weiqiang Zhao, Min-Qi Xue, Xiaogang Du, A. Nandi","doi":"10.3389/frsip.2021.808050","DOIUrl":null,"url":null,"abstract":"Deep convolutional neural networks (DCNNs) have been widely used in medical image segmentation due to their excellent feature learning ability. In these DCNNs, the pooling operation is usually used for image down-sampling, which can gradually reduce the image resolution and thus expands the receptive field of convolution kernel. Although the pooling operation has the above advantages, it inevitably causes information loss during the down-sampling of the pooling process. This paper proposes an effective weighted pooling operation to address the problem of information loss. First, we set up a pooling window with learnable parameters, and then update these parameters during the training process. Secondly, we use weighted pooling to improve the full-scale skip connection and enhance the multi-scale feature fusion. We evaluated weighted pooling on two public benchmark datasets, the LiTS2017 and the CHAOS. The experimental results show that the proposed weighted pooling operation effectively improve network performance and improve the accuracy of liver and liver-tumor segmentation.","PeriodicalId":93557,"journal":{"name":"Frontiers in signal processing","volume":"35 1","pages":""},"PeriodicalIF":1.3000,"publicationDate":"2022-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in signal processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3389/frsip.2021.808050","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

Abstract

Deep convolutional neural networks (DCNNs) have been widely used in medical image segmentation due to their excellent feature learning ability. In these DCNNs, the pooling operation is usually used for image down-sampling, which can gradually reduce the image resolution and thus expands the receptive field of convolution kernel. Although the pooling operation has the above advantages, it inevitably causes information loss during the down-sampling of the pooling process. This paper proposes an effective weighted pooling operation to address the problem of information loss. First, we set up a pooling window with learnable parameters, and then update these parameters during the training process. Secondly, we use weighted pooling to improve the full-scale skip connection and enhance the multi-scale feature fusion. We evaluated weighted pooling on two public benchmark datasets, the LiTS2017 and the CHAOS. The experimental results show that the proposed weighted pooling operation effectively improve network performance and improve the accuracy of liver and liver-tumor segmentation.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
肝脏及肝肿瘤分割池化手术的再思考
深度卷积神经网络(Deep convolutional neural network, DCNNs)由于其出色的特征学习能力,在医学图像分割中得到了广泛的应用。在这些DCNNs中,通常采用池化操作对图像进行下采样,这样可以逐渐降低图像分辨率,从而扩大卷积核的接受域。池化操作虽然具有上述优点,但在池化过程的下采样过程中不可避免地会造成信息丢失。本文提出了一种有效的加权池化操作来解决信息丢失问题。首先,我们建立一个具有可学习参数的池化窗口,然后在训练过程中更新这些参数。其次,利用加权池化方法改进全尺度跳跃连接,增强多尺度特征融合;我们在LiTS2017和CHAOS两个公共基准数据集上评估了加权池化。实验结果表明,所提出的加权池化操作有效地提高了网络性能,提高了肝脏和肝脏肿瘤分割的准确性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
A mini-review of signal processing techniques for RIS-assisted near field THz communication Editorial: Signal processing in computational video and video streaming Editorial: Editor’s challenge—image processing Improved circuitry and post-processing for interleaved fast-scan cyclic voltammetry and electrophysiology measurements Bounds for Haralick features in synthetic images with sinusoidal gradients
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1