arXiv 2404.132222024-05-26eess.IVcs.CV

Vim4Path：用于组织病理学图像的自监督视觉曼巴

Vim4Path: Self-Supervised Vision Mamba for Histopathology Images

Concordia University（康考迪亚大学）；Institute for Research in Immunology and Cancer of the University of Montreal（蒙特利尔大学免疫与癌症研究所）

Ali Nasiri-Sarvi, Vincent Quoc-Huy Trinh, Hassan Rivaz, Mahdi S. Hosseini

备注：: Accepted in CVPR2024 (9th Workshop on Computer Vision for Microscopy Image Analysis)

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出基于视觉曼巴（Vim）的自监督表示学习方法Vim4Path，在Camelyon16上优于ViT，尤其小规模模型ROC AUC提升8.21，且能模拟病理学家工作流程，提升诊断潜力。

摘要：

从千兆像素全切片图像（WSI）中进行表示学习，由于组织结构的复杂性和标记数据的稀缺性，对计算病理学提出了重大挑战。多实例学习方法已应对这一挑战，利用图像块，借助自监督学习（SSL）方法预训练的模型对切片进行分类。SSL和MIL方法的性能均依赖于特征编码器的架构。本文提出在DINO框架内，利用受状态空间模型启发的视觉曼巴（Vim）架构，进行计算病理学中的表示学习。我们在Camelyon16数据集上，对Vim和视觉变换器（ViT）在块级和切片级分类中的性能进行了评估。我们的发现强调了Vim相比ViT的增强性能，尤其是在较小规模下，Vim在相似尺寸模型上的ROC AUC提高了8.21。可解释性分析进一步突出了Vim的能力，揭示了Vim独特地模拟了病理学家的工作流程——与ViT不同。这种与人类专家分析的一致性，突出了Vim在实际诊断环境中的潜力，并为开发计算病理学中有效的表示学习算法做出了重要贡献。我们在\url{https://github.com/AtlasAnalyticsLab/Vim4Path}发布了代码和预训练权重。

英文摘要：

Representation learning from Gigapixel Whole Slide Images (WSI) poses a significant challenge in computational pathology due to the complicated nature of tissue structures and the scarcity of labeled data. Multi-instance learning methods have addressed this challenge, leveraging image patches to classify slides utilizing pretrained models using Self-Supervised Learning (SSL) approaches. The performance of both SSL and MIL methods relies on the architecture of the feature encoder. This paper proposes leveraging the Vision Mamba (Vim) architecture, inspired by state space models, within the DINO framework for representation learning in computational pathology. We evaluate the performance of Vim against Vision Transformers (ViT) on the Camelyon16 dataset for both patch-level and slide-level classification. Our findings highlight Vim's enhanced performance compared to ViT, particularly at smaller scales, where Vim achieves an 8.21 increase in ROC AUC for models of similar size. An explainability analysis further highlights Vim's capabilities, which reveals that Vim uniquely emulates the pathologist workflow-unlike ViT. This alignment with human expert analysis highlights Vim's potential in practical diagnostic settings and contributes significantly to developing effective representation-learning algorithms in computational pathology. We release the codes and pretrained weights at \url{https://github.com/AtlasAnalyticsLab/Vim4Path}.

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。