Triple U-net: Hematoxylin-aware nuclei segmentation with progressive dense feature aggregation

Triple U-net: Hematoxylin-aware nuclei segmentation with progressive dense feature aggregation
复制标题

Triple U-net:具有渐进式密集特征聚合的苏木精感知细胞核分割

DOI:
10.1016/j.media.2020.101786
复制
发表时间:
2020-07
影响因子:
10.9
通讯作者:
Chu Han
Chu Han
中科院分区:
工程技术1区
文献类型:
--
作者:
Bingchao Zhao;Xin Chen;Zhi Li;Zhiwen Yu;Su Yao;Lixu Yan;Yuqian Wang;Zaiyi Liu;Changhong Liang;Chu Han

文献摘要

参考文献

被引文献

相似文献

细胞核分割是肿瘤病理研究的重要环节。由于人工操作的不均匀性、肿瘤细胞核边界模糊、肿瘤细胞重叠等原因,导致彩色图像分割仍是一个有待解决的难题。在本文中,我们的目的是利用独特的光学特性的H&E染色图像,苏木精总是染色细胞核蓝色,伊红总是染色细胞外基质和细胞质粉红色。因此,我们从RGB图像中提取苏木素成分的比尔-朗伯定律。根据光学属性,提取的苏木素组分对颜色不一致具有鲁棒性。利用苏木素成分,我们提出了一个苏木素感知的CNN模型,用于细胞核分割,而无需颜色归一化。我们提出的网络被公式化为一个三重U-网络结构,它包括一个RGB分支,苏木素分支和分割分支。然后,我们提出了一种新的特征聚合策略,允许网络逐步融合特征,并从不同的分支学习更好的特征表示。大量的实验进行定性和定量评估我们所提出的方法的有效性。同时,它在三种不同的细胞核分割数据集上的性能优于最先进的方法。
Nuclei segmentation is a vital step for pathological cancer research. It is still an open problem due to some difficulties, such as color inconsistency introduced by non-uniform manual operations, blurry tumor nucleus boundaries and overlapping tumor cells. In this paper, we aim to leverage the unique optical characteristic of H&E staining images that hematoxylin always stains cell nuclei blue, and eosin always stains the extracellular matrix and cytoplasm pink. Therefore, we extract the Hematoxylin component from RGB images by Beer-Lambert’s Law. According to the optical attribute, the extracted Hematoxylin component is robust to color inconsistency. With the Hematoxylin component, we propose a Hematoxylin-aware CNN model for nuclei segmentation without the necessity of color normalization. Our proposed network is formulated as a Triple U-net structure which includes an RGB branch, a Hematoxylin branch and a Segmentation branch. Then we propose a novel feature aggregation strategy to allow the network to fuse features progressively and to learn better feature representations from different branches. Extensive experiments are performed to qualitatively and quantitatively evaluate the effectiveness of our proposed method. In the meanwhile, it outperforms state-of-the-art methods on three different nuclei segmentation datasets.
DOI: 10.1007/978-3-662-46777-0
发表时间: 2015
期刊: --
影响因子: --
作者:
W. W. Parson-W.
通讯作者: W. W. Parson-W.
DOI: 10.1109/tmi.2018.2865709
发表时间: 2019-02-01
影响因子: 10.6
作者:
Naylor, Peter;Lae, Marick;Walter, Thomas
通讯作者: Walter, Thomas
DOI: 10.1109/tmi.2018.2845918
发表时间: 2018-12-01
影响因子: 10.6
作者:
Li, Xiaomeng;Chen, Hao;Heng, Pheng-Ann
通讯作者: Heng, Pheng-Ann
DOI: --
发表时间: 2019-02
期刊: ArXiv
影响因子: --
作者:
Tien-Ju Yang;Maxwell D. Collins;Yukun Zhu;Jyh-Jing Hwang;Ting Liu;Xiao Zhang;V. Sze;G. Papandreou-G.-Papa
通讯作者: Tien-Ju Yang;Maxwell D. Collins;Yukun Zhu;Jyh-Jing Hwang;Ting Liu;Xiao Zhang;V. Sze;G. Papandreou-G.-Papa
DOI: 10.1109/tmi.2017.2677499
发表时间: 2017-07-01
影响因子: 10.6
作者:
Kumar, Neeraj;Verma, Ruchika;Sethi, Amit
通讯作者: Sethi, Amit