Safdari Reza, Lotfi Dariush, Amiri Mahmood, Peremans Herbert
Medical Biology Research Center, Institute of Health Technology, Kermanshah University of Medical Sciences, Kermanshah, Iran.
Medical Technology Research Center, Institute of Health Technology, Kermanshah University of Medical Sciences, Kermanshah, Iran; Department of Engineering Management, University of Antwerp, Antwerp, Belgium.
Comput Biol Med. 2025 Jul;193:110321. doi: 10.1016/j.compbiomed.2025.110321. Epub 2025 May 24.
This paper presents a novel method for weakly-supervised semantic segmentation (WSSS) of histology images, where only global image-level labels are employed. We leverage an existing weakly-supervised object localization (WSOL) method to generate class activation maps (CAMs) indicating the spatial locations of relevant tissue regions. Next, we utilize a specialized encoder-decoder network to predict fine localization masks. A pixel-wise contrastive loss function is introduced to encourage the model to learn discriminative features for foreground and background regions. Additionally, a pixel-wise cross-entropy loss is incorporated for improved pixel-level supervision. An offline multi-round self-training strategy is also proposed to iteratively refine pseudo masks, enhancing segmentation performance. Our method demonstrates superior segmentation accuracy over the state-of-the-art method on the GlaS dataset (public benchmark for colon cancer). Furthermore, we investigate the efficacy of our approach in a mixed-supervision setting, achieving performance comparable to fully supervised models, indicating its practical applicability in clinical settings. Our results show that the proposed method offers an effective and practical solution for weakly-supervised semantic segmentation in histology images, potentially aiding pathologists in their diagnostic processes and facilitating the development of automated histopathological analysis systems.
本文提出了一种用于组织学图像弱监督语义分割（WSSS）的新方法，该方法仅使用全局图像级标签。我们利用现有的弱监督目标定位（WSOL）方法来生成类激活映射（CAM），以指示相关组织区域的空间位置。接下来，我们使用一个专门的编码器-解码器网络来预测精细的定位掩码。引入了像素级对比损失函数，以鼓励模型学习前景和背景区域的判别特征。此外，还引入了像素级交叉熵损失，以改进像素级监督。还提出了一种离线多轮自训练策略，以迭代地细化伪掩码，提高分割性能。我们的方法在GlaS数据集（结肠癌的公共基准）上展示了优于现有方法的分割精度。此外，我们研究了我们的方法在混合监督设置中的有效性，实现了与完全监督模型相当的性能，表明其在临床环境中的实际适用性。我们的结果表明，所提出的方法为组织学图像中的弱监督语义分割提供了一种有效且实用的解决方案，可能有助于病理学家进行诊断过程，并促进自动组织病理学分析系统的开发。
