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

Engineering human synovial organoids as predictive ecosystems for disease modeling and therapeutic response in rheumatology.

Current opinion in immunology·2026
Same author

Reversing Bladder Cancer Chemo-resistance through Blocking Cell Senescence by a Combination Therapy.

Theranostics·2026
Same author

[Frequency-regulated and multi-scale spatio-temporal modeling-based remote heart rate estimation method].

Sheng wu yi xue gong cheng xue za zhi = Journal of biomedical engineering = Shengwu yixue gongchengxue zazhi·2026
Same author

Evaluating potential effect of adverse childhood experiences on depressive symptoms in Chinese middle-aged and older adults: a simulation-based network analysis.

BMC psychology·2026
Same author

Molecular-Scale Templating to Regulate Active-Rich Closed-Pore Structures in Polycyclic Aromatic Hydrocarbon-Based Hard Carbon for High-Plateau-Capacity Sodium Storage.

ACS applied materials & interfaces·2026
Same author

Regulatory Effects of Paclobutrazol and Uniconazole Mixture on the Morphology and Biomass Allocation of <i>Amorpha fruticosa</i> Seedlings.

Plants (Basel, Switzerland)·2025

相关实验视频

Updated: Jun 23, 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

2.7K

URCA:基于不确定性的区域剪切算法,用于半监督的医疗图像细分.

Chendong Qin1, Yongxiong Wang1, Jiapeng Zhang1

  • 1University of Shanghai for Science and Technology, School of Opto-Electronic Information and Computer Engineering, Department of Control Science and Engineering, 516 War Industrial Road, Shanghai 200093, China.

Computer methods and programs in biomedicine
|June 15, 2024
PubMed
概括
此摘要是机器生成的。

本研究引入了基于不确定性的区域剪切算法,通过提高伪标签质量和解决数据分布偏差来增强半监督医疗图像细分,从而导致更准确的细分结果.

关键词:
分布偏差是一种偏差.医疗图像细分 医疗图像细分非最大的抑制抑制.半监督 半监督 半监督意识到不确定性的不确定性

更多相关视频

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

2.3K
Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.0K

相关实验视频

Last Updated: Jun 23, 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

2.7K
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

2.3K
Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.0K

科学领域:

  • 医学图像分析 医学图像分析
  • 深度学习 (Deep Learning) 是一种深度学习.
  • 计算机视觉 计算机视觉

背景情况:

  • 图像细分的监督学习需要大量的标记数据,这在医学成像中是昂贵和耗时的.
  • 半监督方法显示出希望,但与低可信度的伪标签和标记和未标记数据之间的分布偏差作斗争.

研究的目的:

  • 为了提高半监督医疗图像细分模型的准确性.
  • 解决半监督学习中低信心伪标签和分配偏差的挑战.

主要方法:

  • 提出了一种基于不确定性的区域剪切算法,用于半监督的医疗图像细分.
  • 利用蒙特卡洛掉落来估计不确定性,并采用多种损失函数和非最大抑制来实现模型多样性.
  • 引入了一个新的模块,通过掩盖低可信度像素来生成新样本,增强伪标签的可信度和数据分布.

主要成果:

  • 在 ACDC 和 BraTS2019 基准上比最先进的方法取得了更高的性能.
  • 在ACDC数据集上获得了87.86%的Dice平均得分和4.214毫米的HD95.
  • 达到了大脑瘤细分的84.79%的平均子得分,HD得分为10.13毫米.

结论:

  • 拟议的方法显著提高了半监督医疗图像细分的准确性.
  • 证明了该模型在2D和3D医疗图像数据集上的优势.
  • 代码是公开可用的可复制性和进一步研究.