Han Yongqi, Cheng Lianglun, Huang Guoheng, Zhong Guo, Li Jiahua, Yuan Xiaochen, Liu Hongrui, Li Jiao, Zhou Jian, Cai Muyan
School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, People's Republic of China.
School of Information Science and Technology, Guangdong University of Foreign Studies, Guangzhou 510420, People's Republic of China.
Phys Med Biol. 2023 Feb 7;68(4). doi: 10.1088/1361-6560/acaeee.
. Histopathology image segmentation can assist medical professionals in identifying and diagnosing diseased tissue more efficiently. Although fully supervised segmentation models have excellent performance, the annotation cost is extremely expensive. Weakly supervised models are widely used in medical image segmentation due to their low annotation cost. Nevertheless, these weakly supervised models have difficulty in accurately locating the boundaries between different classes of regions in pathological images, resulting in a high rate of false alarms Our objective is to design a weakly supervised segmentation model to resolve the above problems.. The segmentation model is divided into two main stages, the generation of pseudo labels based on class residual attention accumulation network (CRAANet) and the semantic segmentation based on pixel feature space construction network (PFSCNet). CRAANet provides attention scores for each class through the class residual attention module, while the Attention Accumulation (AA) module overlays the attention feature maps generated in each training epoch. PFSCNet employs a network model containing an inflated convolutional residual neural network and a multi-scale feature-aware module as the segmentation backbone, and proposes dense energy loss and pixel clustering modules are based on contrast learning to solve the pseudo-labeling-inaccuracy problem.. We validate our method using the lung adenocarcinoma (LUAD-HistoSeg) dataset and the breast cancer (BCSS) dataset. The results of the experiments show that our proposed method outperforms other state-of-the-art methods on both datasets in several metrics. This suggests that it is capable of performing well in a wide variety of histopathological image segmentation tasks.. We propose a weakly supervised semantic segmentation network that achieves approximate fully supervised segmentation performance even in the case of incomplete labels. The proposed AA and pixel-level contrast learning also make the edges more accurate and can well assist pathologists in their research.
组织病理学图像分割可以帮助医学专业人员更高效地识别和诊断病变组织。尽管全监督分割模型具有出色的性能，但标注成本极其昂贵。弱监督模型因其标注成本低而被广泛应用于医学图像分割。然而，这些弱监督模型在准确定位病理图像中不同类别区域之间的边界方面存在困难，导致误报率很高。我们的目标是设计一种弱监督分割模型来解决上述问题。该分割模型分为两个主要阶段，即基于类残差注意力积累网络（CRAANet）生成伪标签和基于像素特征空间构建网络（PFSCNet）进行语义分割。CRAANet通过类残差注意力模块为每个类别提供注意力分数，而注意力积累（AA）模块则叠加在每个训练轮次中生成的注意力特征图。PFSCNet采用包含膨胀卷积残差神经网络和多尺度特征感知模块的网络模型作为分割主干，并基于对比学习提出了密集能量损失和像素聚类模块来解决伪标签不准确的问题。我们使用肺腺癌（LUAD-HistoSeg）数据集和乳腺癌（BCSS）数据集对我们的方法进行了验证。实验结果表明，我们提出的方法在这两个数据集的多个指标上均优于其他现有方法。这表明它能够在各种组织病理学图像分割任务中表现出色。我们提出了一种弱监督语义分割网络，即使在标签不完整的情况下也能实现近似全监督的分割性能。所提出的AA和像素级对比学习也使边缘更加准确，并能很好地协助病理学家进行研究。
