FTU-Seek: Foundation Model-Guided Hard-Negative Learning for Sparse Functional Tissue Unit Segmentation
作者: Zonghao Liu, Lei Su, Jiguang Yu, Xuqing Geng, Louis Shuo Wang, Jianmin Wang, Jingfeng Liu
分类: cs.CV, math.NA
发布日期: 2026-09-01
💡 一句话要点
提出FTU-Seek以解决稀疏功能性组织单元分割问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 功能性组织单元 稀疏分割 病理图像分析 负样本选择 深度学习
📋 核心要点
- 现有方法在稀疏功能性组织单元分割中面临挑战,尤其是在形态相似的背景组织干扰下,导致分割精度不足。
- FTU-Seek框架通过利用病理基础模型的特征,采用形态感知的负样本选择策略,提升稀疏FTU的分割效果。
- 通过五折交叉验证和内部测试队列评估,FTU-Seek在TLS、血管和腺体分割任务中表现优异,显著提高了分割精度。
📝 摘要(中文)
功能性组织单元(FTUs),包括三级淋巴结构(TLSs)、血管和腺体,在组织病理学中编码局部免疫、血管和上皮组织。准确量化这些结构对于研究组织架构和与疾病相关的组织组织至关重要。然而,FTUs通常稀疏、异质,并被大量形态相似的背景组织包围,使得在全切片图像(WSIs)中进行自动分割变得具有挑战性。因此,本文开发了FTU-Seek,一个以病理基础模型为指导的框架,将形态感知的负样本选择视为稀疏FTU分割的关键组成部分。FTU-Seek利用来自UNI病理基础模型的冻结多深度特征训练一个补丁级分类器,以区分含FTU和不含FTU的组织。随后,根据预测的目标包含概率对目标缺失补丁进行排名,通过静态Top$K$策略选择得分最高的困难负样本,以构建紧凑的分割训练集。
🔬 方法详解
问题定义:本文旨在解决稀疏功能性组织单元(FTUs)在全切片图像(WSIs)中自动分割的困难,现有方法在处理形态相似的背景组织时表现不佳,导致分割精度低下。
核心思路:FTU-Seek框架的核心思想是利用病理基础模型的多深度特征,通过形态感知的负样本选择来增强稀疏FTU的分割能力,从而提高分类器的准确性。
技术框架:FTU-Seek的整体架构包括特征提取、补丁分类和负样本选择三个主要模块。首先,从UNI病理基础模型中提取冻结的多深度特征;然后训练补丁级分类器以区分含FTU和不含FTU的组织;最后,通过静态Top$K$策略选择困难负样本,构建训练集。
关键创新:FTU-Seek的主要创新在于将形态感知的负样本选择引入稀疏FTU分割任务中,显著提高了模型对稀疏目标的识别能力,与传统方法相比具有本质区别。
关键设计:在设计中,采用了静态Top$K$策略进行负样本选择,确保训练集的紧凑性;同时,使用了适应性损失函数以优化分类器的性能,增强了模型的泛化能力。
🖼️ 关键图片
📊 实验亮点
FTU-Seek在TLS、血管和腺体分割任务中表现出色,经过五折交叉验证,分割精度显著提高,尤其是在与正样本、全组织、随机负样本和匹配随机Top$K$采样策略的比较中,展现出明显的优势。
🎯 应用场景
该研究在病理学领域具有广泛的应用潜力,尤其是在肿瘤微环境分析和疾病相关组织结构研究中。FTU-Seek可以帮助病理学家更准确地识别和量化功能性组织单元,从而为疾病的早期诊断和治疗提供支持。
📄 摘要(原文)
Functional tissue units (FTUs), including tertiary lymphoid structures (TLSs), blood vessels, and glands, encode localized immune, vascular, and epithelial organization in histopathology. Accurate quantification of these structures is important for studying tissue architecture and disease-associated tissue organization. However, FTUs are frequently sparse, heterogeneous, and surrounded by large amounts of morphologically similar background tissue, making automated segmentation in whole-slide images (WSIs) challenging. We therefore developed FTU-Seek, a pathology foundation model-guided framework that treats morphology-aware negative-patch selection as a key component of sparse FTU segmentation. FTU-Seek uses frozen multi-depth features from the UNI pathology foundation model to train a patch-level classifier that distinguishes FTU-containing from FTU-absent tissue. Target-absent patches are subsequently ranked according to their predicted target-containing probabilities, and the highest-scoring hard negatives are selected through a static Top$K$ strategy to construct compact segmentation training sets. The framework was evaluated using five-fold cross-validation and internal test cohorts across TLS, blood-vessel, and gland segmentation tasks, with an additional independent 30-WSI held-out cohort for TLS. Positive-only, all-tissue, random-negative, and matched random Top$K$ sampling strategies served as comparators. Segmentation-derived phenotypes were further explored in external TCGA cohorts.