合成领域定向增强 (S-DOTA) 改善数字病理学模型泛化（英文版）

英文标题：Synthetic DOmain-Targeted Augmentation (S-DOTA) Improves Model Generalization in Digital Pathology中文摘要：本文研究了两种合成模型和基线模型在数字病理学中对模型泛化能力的改进作用，并发现这些方法明显改善了模型在各种任务和设置中的泛化性能，是现实世界应用中应对领域偏移的一种有效方法。英文摘要：Machine learning algorithms have the potential to improve patient outcomes indigital pathology. However, generalization of these tools is currently limitedby sensitivity to variations in tissue preparation, staining procedures andscanning equipment that lead to domain shift in digitized slides. To