SAM-MIL：一种用于全幻灯片图像分类的空间上下文感知多示例学习方法
SAM-MIL: A Spatial Contextual Aware Multiple Instance Learning Approach for Whole Slide Image Classification
AI总结：
针对WSI分类中局部图像块特征提取导致全局空间上下文丢失的问题，提出SAM-MIL框架，利用SAM模型提取图像级空间上下文，并结合组掩码策略与伪袋一致性增强机制，显著提升了分类性能。
摘要：
多示例学习（MIL）是全幻灯片图像（WSI）分类（涵盖亚型分类、诊断等方面）的主要框架。当前的MIL模型主要依赖于从ResNet等预训练模型中提取的实例级特征。这些模型将每张WSI分割为独立的图像块，并从这些局部图像块中提取特征，导致全局空间上下文严重丢失，将模型的注意力限制在局部特征上。为了解决这个问题，我们提出了一种名为SAM-MIL的新型MIL框架，该框架强调空间上下文感知，并通过提取全面的图像级信息来显式地融合空间上下文。Segment Anything Model（SAM）是一种开创性的视觉分割基础模型，能够在无需额外微调的情况下捕获分割特征，这使其成为直接从原始WSI中提取空间上下文的出色工具。我们的方法包括基于空间上下文的组特征提取设计，以及一种用于缓解类别不平衡问题的SAM引导的组掩码策略。我们对不同的分割类别实施动态掩码比例，并用类别的代表性组特征对其进行补充。此外，SAM-MIL通过划分实例来生成额外的伪袋，从而扩充训练集，并引入伪袋之间空间上下文的一致性以进一步提升模型性能。在CAMELYON-16和TCGA肺癌数据集上的实验结果表明，我们提出的SAM-MIL模型在WSI分类任务中优于现有的主流方法。我们的开源实现代码可在 https://github.com/FangHeng/SAM-MIL 获取。
英文摘要：
Multiple Instance Learning (MIL) represents the predominant framework in Whole Slide Image (WSI) classification, covering aspects such as sub-typing, diagnosis, and beyond. Current MIL models predominantly rely on instance-level features derived from pretrained models such as ResNet. These models segment each WSI into independent patches and extract features from these local patches, leading to a significant loss of global spatial context and restricting the model's focus to merely local features. To address this issue, we propose a novel MIL framework, named SAM-MIL, that emphasizes spatial contextual awareness and explicitly incorporates spatial context by extracting comprehensive, image-level information. The Segment Anything Model (SAM) represents a pioneering visual segmentation foundational model that can capture segmentation features without the need for additional fine-tuning, rendering it an outstanding tool for extracting spatial context directly from raw WSIs. Our approach includes the design of group feature extraction based on spatial context and a SAM-Guided Group Masking strategy to mitigate class imbalance issues. We implement a dynamic mask ratio for different segmentation categories and supplement these with representative group features of categories. Moreover, SAM-MIL divides instances to generate additional pseudo-bags, thereby augmenting the training set, and introduces consistency of spatial context across pseudo-bags to further enhance the model's performance. Experimental results on the CAMELYON-16 and TCGA Lung Cancer datasets demonstrate that our proposed SAM-MIL model outperforms existing mainstream methods in WSIs classification. Our open-source implementation code is is available at https://github.com/FangHeng/SAM-MIL.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
