Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
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 JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

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

RstAB activates type 1 fimbriae to promote uropathogenic <i>Escherichia coli</i> bladder invasion.

iScience·2026
Same author

Mechanism Through Which Antioxidant Polysaccharide from <i>Tetrastigma hemsleyanum</i> Protects Against DSS-Induced Ulcerative Colitis: Insights from Multi-Omics.

Molecules (Basel, Switzerland)·2026
Same author

The evolution of liver graft preservation strategies in liver transplantation.

Hepatobiliary surgery and nutrition·2026
Same author

Research progress on the mechanisms and animal models of Ménière's disease.

Genes & diseases·2026
Same author

Extracellular vesicles from chemotherapy-induced senescent tumor cells reprogram hepatic phospholipid metabolism to promote pancreatic cancer liver metastasis.

Journal of experimental & clinical cancer research : CR·2026
Same author

A New Food Supplement Product Containing Ergothioneine Protects Against Cyclophosphamide-Induced Primary Ovarian Insufficiency in Rats.

Molecular nutrition & food research·2026
Same journal

Spatial Coherence Loss: All Objects Matter in Salient and Camouflaged Object Detection.

Pattern recognition·2026
Same journal

LDM-Morph: Latent diffusion model guided deformable image registration.

Pattern recognition·2026
Same journal

Variable Priority for Unsupervised Variable Selection.

Pattern recognition·2026
Same journal

A Deep Spatio-Temporal Architecture for Dynamic ECN Analysis with Granger Causality based Causal Discovery.

Pattern recognition·2025
Same journal

Medical image segmentation using dual-decoder mutual teaching with a mean teacher framework.

Pattern recognition·2025
Same journal

Multi-graph Graph matching for coronary artery semantic labeling in invasive coronary angiograms.

Pattern recognition·2025
查看所有相关文章

相关实验视频

Updated: Jul 15, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.8K

从多个注释器学习医疗图像细分的学习.

Le Zhang1,2, Ryutaro Tanno3, Moucheng Xu2

  • 1Queen Square Institute of Neurology, Faculty of Brain Sciences, University College London, London, WC1B 5EH, United Kingdom.

Pattern recognition
|October 2, 2023
PubMed
概括
此摘要是机器生成的。

这项研究引入了一种新的深度学习方法,通过考虑注释器的可靠性和共识来改善医疗图像细分. 这种方法提高了细分的准确性,特别是在有限或冲突的数据.

关键词:
标签融合 标签融合 标签融合多个注释器的多个注释器分段化 分段化 分段化 分段化

更多相关视频

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

439
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.5K

相关实验视频

Last Updated: Jul 15, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

439
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.5K

科学领域:

  • 医学图像分析 医学图像分析
  • 机器学习 机器学习
  • 计算机视觉 计算机视觉

背景情况:

  • 用于细分的监督机器学习对标签质量很敏感.
  • 医学图像注释受高成本和观察者之间的变化影响.
  • 噪音标签限制了自动细分算法的性能.

研究的目的:

  • 开发一种方法,从杂的观测中共同学习注释器可靠性和专家共识标签.
  • 为了提高医疗图像细分精度,尽管标签噪声和分歧.
  • 为了应对医疗成像中的高成本,可变注释的挑战.

主要方法:

  • 使用了两个合的卷积神经网络 (CNN).
  • 通过单独从噪音数据中共同学习注释器可靠性和共识标签分布.
  • 采用"最大限度不可靠"的策略,将注释者行为与共识分开.
  • 在MNIST,ISBI2015,BraTS,LIDC-IDRI和一个新的QSMSC数据集上进行验证.

主要成果:

  • 建议的方法在所有测试的数据集中始终优于竞争方法和基线.
  • 性能增长最显著的是小数量的注释和高分歧.
  • 该系统有效地捕获了注释器错误的复杂空间特征.
  • 在各种医学成像任务上表现出有效性,包括多发性硬化病变,脑瘤和肺部异常.

结论:

  • 这种新的CNN方法成功地解决了医疗图像细分中的噪音和可变标签的挑战.
  • 这种方法提供了一个强大的解决方案,可以在数据稀缺或高不一致的场景中提高细分精度.
  • 模拟注释器可靠性的能力提高了自动化细分系统的可靠性和性能.