Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Contrasting Leaf Adaptation Strategies of <i>Pinus koraiensis</i> and <i>Fraxinus mandshurica</i> Under Water and Nutrient Variation.

Plants (Basel, Switzerland)·2026

Same author

Ascending aortic versus axillary artery cannulation in Stanford type A aortic dissection repair: a propensity score-weighted cohort study.

Journal of thoracic disease·2026

Same author

Momentum contrast learning-based multimodal digital content infringement detection.

Scientific reports·2026

Same author

The Dual Roles of Innate Lymphoid Cells in Antibacterial Defense and Bacterial Immune Evasion.

Clinical reviews in allergy & immunology·2026

Same author

Cultivar and tree age jointly shape rhizosphere bacterial diversity and community assembly in Macadamia integrifolia.

Applied microbiology and biotechnology·2026

Same author

<i>Escherichia coli JNL-EC1</i> enhances type I IFN-mediated antiviral response during DNA and RNA virus infection.

Frontiers in microbiology·2026

Search research articles

相关实验视频

Updated: Jun 25, 2025

Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

Published on: July 26, 2014

面具超图学习对弱监督的组织病理学全幻灯片图像分类.

Jun Shi¹, Tong Shu², Kun Wu³

¹School of Software, Hefei University of Technology, Hefei, 230601, Anhui Province, China.

Computer methods and programs in biomedicine

|May 31, 2024

概括

这项研究介绍了蒙面超图学习 (MaskHGL) 通过捕捉图像补丁之间的复杂,非对式关系来分析全幻灯片图像 (WSI) 的基因病理学. 这种新的方法显著提高了WSI分类的准确性和稳定性,显示了癌症亚型和基因突变预测的前景.

关键词:

计算病理学计算病理学计算机辅助诊断是一种计算机辅助的诊断.超图形学习的学习方法监督的弱点监督的弱点整个幻灯片图像的分类整体幻灯片图像的分类.

更多相关视频

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research

Published on: June 21, 2024

相关实验视频

Last Updated: Jun 25, 2025

Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

Published on: July 26, 2014

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research

Published on: June 21, 2024

科学领域:

计算病理学计算病理学
机器学习用于医学成像.
图形神经网络是一个神经网络.

背景情况:

整个幻灯片图像 (WSI) 在组织病理学中提出了复杂的数据挑战.
现有的图形神经网络 (GNN) 方法用于WSI分析,往往忽略了非对式补丁关系.
这种限制阻碍了最佳的特征学习和分类性能.

研究的目的:

探索和利用在组织病理学WSIs中的非对式关系.
开发一个新的框架来学习幻灯片级别的表示.
通过整合非对式补丁相关性来提高WSI分类性能.

主要方法:

提出了一个掩盖的超图学习 (MaskHGL) 框架,用于弱监督的WSI分类.
利用超图来建模非对式补丁相关性,并使用超图卷积来实现全球消息传递.
整合了一个掩盖的超图重建模块,以提高稳定性和概括性,以及一个自我注意节点聚合器.

主要成果:

在TCGA-LUNG,TCGA-EGFR和USTC-EGFR数据集上评估了MaskHGL.
达到高的ROC曲线下的面积 (AUC) 值:0.9752±0.0024 (TCGA-LUNG),0.7421±0.0380 (TCGA-EGFR) 和0.8745±0.0100 (USTC-EGFR). 在这个过程中,我们可以看到ROC曲线下的面积.
在最先进的方法中表现优越,在USTC-EGFR数据集上SlideGraph+的表现优于2.64%.

结论:

MaskHGL通过利用非对式关系显著改善了WSI分类.
蒙面超图重建模块提高了稳定性和分类能力,特别是在数据稀缺的场景中.
该方法显示了癌症亚型的强大潜力,并预测了来自H&E染色WSIs的基因突变.