Annu Int Conf IEEE Eng Med Biol Soc. 2021 Nov;2021:3447-3450. doi: 10.1109/EMBC46164.2021.9630818.
Histopathology digital scans are large-size images that contain valuable information at the pixel level. Contentbased comparison of these images is a challenging task. This study proposes a content-based similarity measure for highresolution gigapixel histopathology images. The proposed similarity measure is an expansion of cosine vector similarity to a matrix. Each image is divided into same-size patches with a meaningful amount of information (i.e., contained enough tissue). The similarity is measured by the extraction of patchlevel deep embeddings of the last pooling layer of a pre-trained deep model at four different magnification levels, namely, 1x, 2.5x, 5x, and 10x magnifications. In addition, for faster measurement, embedding reduction is investigated. Finally, to assess the proposed method, an image search method is implemented. Results show that the similarity measure represents the slide labels with a maximum accuracy of 93.18% for top-5 search at 5x magnification.
组织病理学数字扫描是包含像素级有价值信息的大尺寸图像。对这些图像进行基于内容的比较是一项具有挑战性的任务。本研究提出了一种用于高分辨率千兆像素组织病理学图像的基于内容的相似性度量方法。所提出的相似性度量是余弦向量相似性到矩阵的扩展。每个图像都被划分为具有一定信息量的相同大小的斑块（即包含足够的组织）。通过在四个不同的放大倍数水平（即 1x、2.5x、5x 和 10x 放大倍数）提取预训练深度模型的最后池化层的斑块级深度嵌入来测量相似性。此外，为了加快测量速度，还研究了嵌入缩减。最后，为了评估所提出的方法，实现了一种图像搜索方法。结果表明，相似性度量方法能够以 5x 放大倍数下的最高准确率 93.18%表示幻灯片标签，用于前 5 名搜索。
