arXiv 2302.098332023-05-03cs.CVcs.AIcs.LG

领域特定预训练提升全切片图像分类的置信度

Domain-Specific Pre-training Improves Confidence in Whole Slide Image Classification

Birla Institute of Technology and Science, Pilani（毕拉理工与科学大学 Pilani校区）；Macquarie University（麦考瑞大学）；University of California, San Diego（加利福尼亚大学圣地亚哥分校）；TCS Research（塔塔咨询服务公司研究院）

Soham Rohit Chitnis, Sidong Liu, Tirtharaj Dash, Tanmay Tulsidas Verlekar, Antonio Di Ieva, Shlomo Berkovsky, Lovekesh Vig, Ashwin Srinivasan

备注：: Accepted in EMBC 2023
发表信息：: Annu Int Conf IEEE Eng Med Biol Soc (EMBC 2023)
DOI：: 10.1109/EMBC40787.2023.10340659

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文研究领域特定预训练对WSI分类的影响，通过在CLAM和TransMIL模型上的实验，证明其在胶质瘤亚型分类中提升了模型置信度并达到新最优性能。

摘要：

全切片图像（WSIs）或组织病理学图像被用于数字病理学。由于WSIs的尺寸巨大且缺乏像素级标注，它们给用于临床诊断的深度学习模型带来了巨大挑战。随着计算病理学的最新进展，人们提出了更新的基于多示例学习的模型。针对WSIs的多示例学习需要创建图块，并使用这些图块的编码进行诊断。这些模型使用通用预训练模型（在ImageNet上预训练的ResNet-50）进行图块编码。最近提出的KimiaNet是一个在TCGA切片上预训练的DenseNet121模型，属于领域特定预训练模型。本文展示了领域特定预训练对WSI分类的影响。为了研究领域特定预训练的效果，我们考虑了当前最先进的两种多示例学习模型：1）基于注意力的模型CLAM，以及2）基于自注意力的模型TransMIL，并评估了这些模型在检测原发性脑肿瘤——胶质瘤时的置信度和预测性能。领域特定预训练提升了模型的置信度，并在基于WSI的胶质瘤亚型分类中达到了新的最先进性能，显示出在辅助胶质瘤诊断方面的高临床适用性。我们将公开分享我们的代码和实验结果。

英文摘要：

Whole Slide Images (WSIs) or histopathology images are used in digital pathology. WSIs pose great challenges to deep learning models for clinical diagnosis, owing to their size and lack of pixel-level annotations. With the recent advancements in computational pathology, newer multiple-instance learning-based models have been proposed. Multiple-instance learning for WSIs necessitates creating patches and uses the encoding of these patches for diagnosis. These models use generic pre-trained models (ResNet-50 pre-trained on ImageNet) for patch encoding. The recently proposed KimiaNet, a DenseNet121 model pre-trained on TCGA slides, is a domain-specific pre-trained model. This paper shows the effect of domain-specific pre-training on WSI classification. To investigate the effect of domain-specific pre-training, we considered the current state-of-the-art multiple-instance learning models, 1) CLAM, an attention-based model, and 2) TransMIL, a self-attention-based model, and evaluated the models' confidence and predictive performance in detecting primary brain tumors - gliomas. Domain-specific pre-training improves the confidence of the models and also achieves a new state-of-the-art performance of WSI-based glioma subtype classification, showing a high clinical applicability in assisting glioma diagnosis. We will publicly share our code and experimental results at https://github.com/soham-chitnis10/WSI-domain-specific.

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。