排名损失和隔离学习用于降低组织病理学图像搜索偏差（英文版）

英文标题：Ranking Loss and Sequestering Learning for Reducing Image Search Bias in Histopathology中文摘要：本文提出两个新思想来提高医疗图像搜索性能，采用排名损失函数引导特征提取，将表示学习定制为图像搜索而不是学习类标签，同时引入了隔离学习的概念来增强特征提取的泛化性能，并通过最大的公共数据集实现验证，实验结果与现有技术相比具有更好的表现。英文摘要：Recently, deep learning has started to play an essential role in healthcareapplications, including image search in digital pathology. Despite the recentprogress in computer vision, significant issues remain for image searching inhistopathology archives. A well-known problem is AI