BriefGPT.xyz
Jan, 2024
基础模型的低资源化微调在组织病理学中超越了现有技术水平
Low-resource finetuning of foundation models beats state-of-the-art in histopathology
HTML
PDF
Benedikt Roth, Valentin Koch, Sophia J. Wagner, Julia A. Schnabel, Carsten Marr...
TL;DR
通过细化基础模型,仅经历两小时或三天的单个GPU训练,我们可以在计算病理学中的特征提取上相媲美或超越现有的方法,这表示即使资源有限,也可以针对特定下游任务和数据集训练一个定制的特征提取器。
Abstract
To handle the large scale of
whole slide images
in
computational pathology
, most approaches first tessellate the images into smaller patches, extract features from these patches, and finally aggregate the feature
→