分层图像金字塔Transformer在组织病理学图像中结肠活检和息肉分类中的基准测试
Benchmarking Hierarchical Image Pyramid Transformer for the classification of colon biopsies and polyps in histopathology images
AI总结：
本研究基准测试分层图像金字塔Transformer（HIPT）在结直肠活检和息肉分类中的表现，比较基于TCGA权重微调与随机初始化预训练两种策略在二分类和多分类任务上的效果。
摘要：
在组织病理学全切片图像（WSI）中，由于WSI具有十亿像素级分辨率，使用高质量像素级标注训练神经网络是一个昂贵的过程。然而，自监督学习的最新进展表明，无需标注即可学习到高度描述性的图像表示。我们研究了最新的分层图像金字塔Transformer（HIPT）模型在结直肠活检和息肉分类这一特定任务中的应用。在评估了原始HIPT模型中TCGA学习特征的有效性后，我们使用两种不同策略将结肠活检图像信息纳入HIPT的预训练：（1）从现有TCGA权重对HIPT进行微调；（2）从随机权重初始化对HIPT进行预训练。我们在两个结直肠活检分类任务上比较了这些预训练方案的性能：二分类和多分类。
英文摘要：
Training neural networks with high-quality pixel-level annotation in histopathology whole-slide images (WSI) is an expensive process due to gigapixel resolution of WSIs. However, recent advances in self-supervised learning have shown that highly descriptive image representations can be learned without the need for annotations. We investigate the application of the recent Hierarchical Image Pyramid Transformer (HIPT) model for the specific task of classification of colorectal biopsies and polyps. After evaluating the effectiveness of TCGA-learned features in the original HIPT model, we incorporate colon biopsy image information into HIPT's pretraining using two distinct strategies: (1) fine-tuning HIPT from the existing TCGA weights and (2) pretraining HIPT from random weight initialization. We compare the performance of these pretraining regimes on two colorectal biopsy classification tasks: binary and multiclass classification.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
