Robust Histopathology Image Analysis: to Label or to Synthesize?
Robust Histopathology Image Analysis: to Label or to Synthesize?
复制标题
DOI：
10.1109/cvpr.2019.00873
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Saltz JH
中科院分区：
文献类型：
--
作者：
Hou L;Agarwal A;Samaras D;Kurc TM;Gupta RR;Saltz JH
细胞核的检测、分割和分类是数字病理学中的基本分析操作。现有的最先进的方法需要来自病理学家的大量监督训练数据，并且在来自看不见的组织类型的图像中可能仍然表现不佳。我们提出了一种无监督的方法，组织病理学图像分割，合成异构集的训练图像补丁，每种组织类型。虽然我们的合成补丁并不总是高质量的，我们通过一个普遍适用的重要性抽样方法利用生成的样本的杂色船员。这种提出的方法首次重新权衡合成数据的训练损失，以便最大限度地减少真实数据分布的理想（无偏）泛化损失。这使得我们能够使用随机多边形生成器来合成近似的细胞结构（即，核掩模），在许多组织类型中没有给出真实的例子，因此，基于GAN的方法不适合。此外，我们提出了一个混合合成管道，利用纹理在真实的组织病理学补丁和GAN模型，以解决组织纹理的异质性。与现有的最先进的监督模型相比，我们的方法在没有训练数据的情况下对癌症类型的泛化效果明显更好。即使在具有训练数据的癌症类型中，我们的方法也可以在没有监督成本的情况下实现相同的性能。我们在癌症基因组图谱（TCGA）存储库中发布了超过5000个全载玻片图像（WSI）的代码和分割结果1，该数据集将比目前可用的数据集大几个数量级。
Detection, segmentation and classification of nuclei are fundamental analysis operations in digital pathology. Existing state-of-the-art approaches demand extensive amount of supervised training data from pathologists and may still perform poorly in images from unseen tissue types. We propose an unsupervised approach for histopathology image segmentation that synthesizes heterogeneous sets of training image patches, of every tissue type. Although our synthetic patches are not always of high quality, we harness the motley crew of generated samples through a generally applicable importance sampling method. This proposed approach, for the first time, re-weighs the training loss over synthetic data so that the ideal (unbiased) generalization loss over the true data distribution is minimized. This enables us to use a random polygon generator to synthesize approximate cellular structures (i.e., nuclear masks) for which no real examples are given in many tissue types, and hence, GAN-based methods are not suited. In addition, we propose a hybrid synthesis pipeline that utilizes textures in real histopathology patches and GAN models, to tackle heterogeneity in tissue textures. Compared with existing state-of-the-art supervised models, our approach generalizes significantly better on cancer types without training data. Even in cancer types with training data, our approach achieves the same performance without supervision cost. We release code and segmentation results1 on over 5000 Whole Slide Images (WSI) in The Cancer Genome Atlas (TCGA) repository, a dataset that would be orders of magnitude larger than what is available today.