Yu Jiahui, Ma Tianyu, Hua Dong, Chen Feng, Fu Junfen, Xu Yingke
IEEE J Biomed Health Inform. 2024 Jul 31;PP. doi: 10.1109/JBHI.2024.3436099.
Current whole slide image (WSI) segmentation aims at extracting tumor regions from the background. Unlike this, segmenting distinct tumor areas (instances) within a WSI driven by limited annotated data remains under-explored. In this paper, we formally propose semisupervised instance segmentation (Semi-IS) in WSIs. We address a key challenge: learning intra-class similarity and inter-class dissimilarity driven by unlabeled data. Specifically, we generally perceive the patch as composed of tokens (together), not the patch alone. We employ contrastive learning to develop a segmentation framework. In the SemiIS, we find that the boundaries of segmented instances are usually disturbed by noise. We jointly eliminate and preserve noise features to address this problem. We conduct extensive experiments to evaluate the effectiveness and generalizability of Semi-IS, including histopathology and cellular pathology. The results show that in clinical multi instance segmentation tasks, Semi-IS achieves almost fullsupervised state-of-the-art results with only 30% annotated data. Semi-IS can improve segmentation accuracy by about 2% on public cell pathology datasets.
当前的全切片图像（WSI）分割旨在从背景中提取肿瘤区域。与此不同的是，在有限标注数据驱动下对WSI内不同的肿瘤区域（实例）进行分割仍未得到充分探索。在本文中，我们正式提出了WSI中的半监督实例分割（Semi-IS）。我们解决了一个关键挑战：学习由未标注数据驱动的类内相似性和类间差异性。具体而言，我们通常将图像块视为由多个标记（共同）组成，而不是单独的图像块。我们采用对比学习来开发一个分割框架。在SemiIS中，我们发现分割实例的边界通常会受到噪声干扰。我们联合消除并保留噪声特征来解决这个问题。我们进行了广泛的实验来评估Semi-IS的有效性和通用性，包括组织病理学和细胞病理学。结果表明，在临床多实例分割任务中，Semi-IS仅使用30%的标注数据就取得了几乎与完全监督相当的最优结果。在公共细胞病理学数据集上，Semi-IS可以将分割准确率提高约2%。
