Sajjad Usama, Rezapour Mostafa, Su Ziyu, Tozbikian Gary H, Gurcan Metin N, Niazi M Khalid Khan
Center for Biomedical Informatics, Wake Forest University School of Medicine, Winston-Salem, NC 27101, USA.
Department of Pathology, The Ohio State University, Columbus, OH 43210, USA.
Cancers (Basel). 2023 Jun 30;15(13):3428. doi: 10.3390/cancers15133428.
The early diagnosis of lymph node metastasis in breast cancer is essential for enhancing treatment outcomes and overall prognosis. Unfortunately, pathologists often fail to identify small or subtle metastatic deposits, leading them to rely on cytokeratin stains for improved detection, although this approach is not without its flaws. To address the need for early detection, multiple-instance learning (MIL) has emerged as the preferred deep learning method for automatic tumor detection on whole slide images (WSIs). However, existing methods often fail to identify some small lesions due to insufficient attention to small regions. Attention-based multiple-instance learning (ABMIL)-based methods can be particularly problematic because they may focus too much on normal regions, leaving insufficient attention for small-tumor lesions. In this paper, we propose a new ABMIL-based model called normal representative keyset ABMIL (NRK-ABMIL), which addresseses this issue by adjusting the attention mechanism to give more attention to lesions. To accomplish this, the NRK-ABMIL creates an optimal keyset of normal patch embeddings called the normal representative keyset (NRK). The NRK roughly represents the underlying distribution of all normal patch embeddings and is used to modify the attention mechanism of the ABMIL. We evaluated NRK-ABMIL on the publicly available Camelyon16 and Camelyon17 datasets and found that it outperformed existing state-of-the-art methods in accurately identifying small tumor lesions that may spread over a few patches. Additionally, the NRK-ABMIL also performed exceptionally well in identifying medium/large tumor lesions.
乳腺癌淋巴结转移的早期诊断对于提高治疗效果和总体预后至关重要。不幸的是，病理学家常常难以识别微小或不明显的转移灶，这使得他们依赖细胞角蛋白染色来提高检测率，尽管这种方法并非毫无缺陷。为满足早期检测的需求，多实例学习（MIL）已成为在全切片图像（WSIs）上进行自动肿瘤检测的首选深度学习方法。然而，由于对小区域关注不足，现有方法常常无法识别一些小病变。基于注意力的多实例学习（ABMIL）方法可能尤其成问题，因为它们可能过于关注正常区域，而对小肿瘤病变的关注不足。在本文中，我们提出了一种新的基于ABMIL的模型，称为正常代表性键集ABMIL（NRK-ABMIL），它通过调整注意力机制以更多地关注病变来解决这个问题。为实现这一点，NRK-ABMIL创建了一个称为正常代表性键集（NRK）的正常补丁嵌入的最优键集。NRK大致代表了所有正常补丁嵌入的潜在分布，并用于修改ABMIL的注意力机制。我们在公开可用的Camelyon16和Camelyon17数据集上评估了NRK-ABMIL，发现它在准确识别可能分布在几个补丁上的小肿瘤病变方面优于现有的最先进方法。此外，NRK-ABMIL在识别中/大肿瘤病变方面也表现出色。
