HookMIL: 重新审视多实例学习中的上下文建模以用于计算病理学
HookMIL: Revisiting Context Modeling in Multiple Instance Learning for Computational Pathology
AI总结：
HookMIL通过引入可学习的钩子标记和多样性损失，提升多实例学习在计算病理学中的效率与可解释性。
摘要：
多实例学习（MIL）已使计算病理学中对整个滑动玻璃片（WSI）的弱监督分析成为可能。然而，传统MIL方法往往丢失关键的上下文信息，而基于变压器的变体虽然更具表现力，但存在二次复杂度和冗余计算的问题。为了解决这些限制，我们提出了HookMIL，一种具有上下文意识且计算高效的MIL框架，该框架利用紧凑且可学习的钩子标记进行结构化的上下文聚合。这些标记可以初始化自（i）关键-补丁视觉特征，（ii）来自视觉-语言病理学模型的文本嵌入，以及（iii）来自空间转录组-视觉模型的空间性特征。这种多模态初始化使钩子标记能够融合丰富的文本和空间先验，加速收敛并提高表示质量。在训练过程中，钩子标记通过双向注意力机制与实例交互，具有线性复杂度。为进一步促进专业化，我们引入了钩子多样性损失，鼓励每个标记专注于不同的组织病理学模式。此外，钩子到钩子的通信机制通过最小化冗余来细化上下文交互。在四个公开病理学数据集上的广泛实验表明，HookMIL实现了最先进的性能，具有改进的计算效率和可解释性。代码可在https://github.com/lingxitong/HookMIL上获得。
英文摘要：
Multiple Instance Learning (MIL) has enabled weakly supervised analysis of whole-slide images (WSIs) in computational pathology. However, traditional MIL approaches often lose crucial contextual information, while transformer-based variants, though more expressive, suffer from quadratic complexity and redundant computations. To address these limitations, we propose HookMIL, a context-aware and computationally efficient MIL framework that leverages compact, learnable hook tokens for structured contextual aggregation. These tokens can be initialized from (i) key-patch visual features, (ii) text embeddings from vision-language pathology models, and (iii) spatially grounded features from spatial transcriptomics-vision models. This multimodal initialization enables Hook Tokens to incorporate rich textual and spatial priors, accelerating convergence and enhancing representation quality. During training, Hook tokens interact with instances through bidirectional attention with linear complexity. To further promote specialization, we introduce a Hook Diversity Loss that encourages each token to focus on distinct histopathological patterns. Additionally, a hook-to-hook communication mechanism refines contextual interactions while minimizing redundancy. Extensive experiments on four public pathology datasets demonstrate that HookMIL achieves state-of-the-art performance, with improved computational efficiency and interpretability. Codes are available at https://github.com/lingxitong/HookMIL.
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。
