重新思考基于注意力的多实例学习用于全切片病理图像分类：实例属性视角
Rethinking Attention-Based Multiple Instance Learning for Whole-Slide Pathological Image Classification: An Instance Attribute Viewpoint
AI总结：
本文提出属性驱动的多实例学习框架AttriMIL，通过属性评分机制及空间与排序约束建模实例相关性，并采用病理自适应主干网络，在三个公开基准上超越现有方法。
摘要：
多实例学习（MIL）是全切片病理图像（WSI）分析的一种稳健范式，利用切片级标签处理十亿像素级分辨率的图像。作为开创性工作，基于注意力的MIL（ABMIL）及其变体由于能够同时处理临床诊断和肿瘤定位而日益流行。然而，注意力机制在区分实例方面存在局限性，常常导致组织误分类，并可能损害MIL性能。本文提出了一种属性驱动的MIL（AttriMIL）框架来解决这些问题。具体而言，我们剖析了ABMIL的计算过程，并提出了一种属性评分机制，该机制有效衡量每个实例对包预测的贡献，从而量化实例属性。基于属性量化，我们开发了空间属性约束和属性排序约束，分别用于建模切片内部和切片之间的实例相关性。这些约束促使网络捕获实例的空间相关性和语义相似性，提高AttriMIL区分组织类型和识别具有挑战性的实例的能力。此外，AttriMIL采用了一种组织病理学自适应主干网络，最大化预训练模型的特征提取能力以收集病理特征。在三个公开基准上的大量实验表明，我们的AttriMIL在多个评估指标上优于现有的最先进框架。实现代码可在https://github.com/MedCAI/AttriMIL获取。
英文摘要：
Multiple instance learning (MIL) is a robust paradigm for whole-slide pathological image (WSI) analysis, processing gigapixel-resolution images with slide-level labels. As pioneering efforts, attention-based MIL (ABMIL) and its variants are increasingly becoming popular due to the characteristics of simultaneously handling clinical diagnosis and tumor localization. However, the attention mechanism exhibits limitations in discriminating between instances, which often misclassifies tissues and potentially impairs MIL performance. This paper proposes an Attribute-Driven MIL (AttriMIL) framework to address these issues. Concretely, we dissect the calculation process of ABMIL and present an attribute scoring mechanism that measures the contribution of each instance to bag prediction effectively, quantifying instance attributes. Based on attribute quantification, we develop a spatial attribute constraint and an attribute ranking constraint to model instance correlations within and across slides, respectively. These constraints encourage the network to capture the spatial correlation and semantic similarity of instances, improving the ability of AttriMIL to distinguish tissue types and identify challenging instances. Additionally, AttriMIL employs a histopathology adaptive backbone that maximizes the pre-trained model's feature extraction capability for collecting pathological features. Extensive experiments on three public benchmarks demonstrate that our AttriMIL outperforms existing state-of-the-art frameworks across multiple evaluation metrics. The implementation code is available at https://github.com/MedCAI/AttriMIL.
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。
