利用注意力机制蒸馏高诊断价值图像块的全切片图像分类方法
Distilling High Diagnostic Value Patches for Whole Slide Image Classification Using Attention Mechanism
AI总结：
本文提出AFD-MIL方法，通过注意力机制蒸馏高诊断价值图像块并排除冗余块，结合全局损失优化，在Camelyon16和TCGA-NSCLC上取得优于现有方法的分类性能。
摘要：
多实例学习（MIL）在全切片图像（WSI）分类领域受到广泛关注，因为它用诊断报告作为标签替代像素级人工标注，显著降低了人力成本。近期研究表明，包级MIL方法通常能取得更好的结果，因为它们可以将WSI的所有图像块作为一个整体来考虑。然而，这类方法的一个缺点是纳入了更多冗余图像块，从而导致干扰。为了提取具有高诊断价值的图像块并排除干扰图像块以解决这一问题，我们开发了一种基于注意力的特征蒸馏多实例学习（AFD-MIL）方法。该方法提出将排除冗余图像块作为弱监督学习中的预处理操作，直接减轻大量噪声带来的干扰。它还开创性地使用注意力机制来蒸馏具有高诊断价值的特征，不同于传统做法中不加区分地强制整合所有图像块。此外，我们引入了全局损失优化来精细控制特征蒸馏模块。AFD-MIL与许多现有MIL方法正交，因此能带来一致的性能提升。该方法已超越当前最先进的方法，在Camelyon16（Camelyon Challenge 2016，乳腺癌）上取得了91.47%的ACC（准确率）和94.29%的AUC（曲线下面积），在TCGA-NSCLC（癌症基因组图谱计划：非小细胞肺癌）上取得了93.33%的ACC和98.17%的AUC。针对两个数据集使用了不同的特征蒸馏方法，以适应特定疾病，从而提高了性能和可解释性。
英文摘要：
Multiple Instance Learning (MIL) has garnered widespread attention in the field of Whole Slide Image (WSI) classification as it replaces pixel-level manual annotation with diagnostic reports as labels, significantly reducing labor costs. Recent research has shown that bag-level MIL methods often yield better results because they can consider all patches of the WSI as a whole. However, a drawback of such methods is the incorporation of more redundant patches, leading to interference. To extract patches with high diagnostic value while excluding interfering patches to address this issue, we developed an attention-based feature distillation multi-instance learning (AFD-MIL) approach. This approach proposed the exclusion of redundant patches as a preprocessing operation in weakly supervised learning, directly mitigating interference from extensive noise. It also pioneers the use of attention mechanisms to distill features with high diagnostic value, as opposed to the traditional practice of indiscriminately and forcibly integrating all patches. Additionally, we introduced global loss optimization to finely control the feature distillation module. AFD-MIL is orthogonal to many existing MIL methods, leading to consistent performance improvements. This approach has surpassed the current state-of-the-art method, achieving 91.47% ACC (accuracy) and 94.29% AUC (area under the curve) on the Camelyon16 (Camelyon Challenge 2016, breast cancer), while 93.33% ACC and 98.17% AUC on the TCGA-NSCLC (The Cancer Genome Atlas Program: non-small cell lung cancer). Different feature distillation methods were used for the two datasets, tailored to the specific diseases, thereby improving performance and interpretability.
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。
