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

Rising Global Burden and Sex Disparities in Nonmelanoma Skin Cancer: 1990-2021 Trends With Projections to 2036.

Advances in skin & wound care·2026
Same author

Novel Dual Soft Drug Strategy Enables Development of Topical Androgen Receptor Antagonists with Enhanced Efficacy and Optimized Safety for Androgenetic Alopecia.

Journal of medicinal chemistry·2026
Same author

Variations in GHG fluxes in small- and medium-sized water bodies in different climate zones.

Journal of environmental management·2026
Same author

Thermo-Mechanical Degradation Behavior of the Base-Subgrade Interface in Airport Pavements: A Sequentially Coupled Cohesive-Zone Study.

Materials (Basel, Switzerland)·2026
Same author

Nitrate contamination characteristics and health risk assessment of groundwater in the typical area of the lower Yellow River Basin.

PloS one·2026
Same author

Synthetic microbial communities: a novel emerging models for dissecting gut microbiota-host interactions in neurodegenerative diseases.

Frontiers in immunology·2026

相关实验视频

Updated: Jul 3, 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

基于对应的生成贝叶斯深度学习用于半监督的体积医学图像细分.

Yuzhou Zhao1, Xinyu Zhou1, Tongxin Pan1

  • 1Shanghai Key Lab of Intelligent Information Processing, School of Computer Science, Fudan University, Shanghai, China.

Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society
|February 11, 2024
PubMed
概括

这项研究引入了一种新的基于对应的生成贝叶斯深度学习 (C-GBDL) 模型,用于半监督的医疗图像细分. C-GBDL模型提高了伪标签质量,提高了在有限的标签数据下细分的准确性.

关键词:
贝叶斯的深度学习是贝叶斯的深度学习.双重不确定性估计估计医疗图像细分 医疗图像细分语义对应的语义对应.半监督 半监督 半监督 半监督 半监督

更多相关视频

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

402

相关实验视频

Last Updated: Jul 3, 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
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

402

科学领域:

  • 医疗图像分析 医学图像分析
  • 计算机辅助诊断是一种计算机辅助的诊断.
  • 机器学习在医疗保健中的应用

背景情况:

  • 自动化医疗图像细分对于临床应用至关重要.
  • 高昂的注释成本限制了完全监督的方法,推动了对半监督方法的兴趣.
  • 现有的半监督方法由于数据分布偏差而难以获得伪标签质量.

研究的目的:

  • 引入一种创新的基于对应的生成贝叶斯深度学习 (C-GBDL) 模型.
  • 改善在半监督医疗细分中生成高质量的伪标签.
  • 为了解决数据分布偏差,提高细分精度.

主要方法:

  • 开发了一个教师-学生架构,结合了多层次的语义对应方法.
  • 教师模型通过与参考卷相匹配的特征来学习概括的数据分布.
  • 提出了一种双重不确定性估计方案 (预测和结构相似性),以改进伪标签.

主要成果:

  • 在比较实验中,C-GBDL模型表现出卓越的性能.
  • 在两个公共医疗数据集上验证了有效性.
  • 与现有的半监督方法相比,实现了更好的细分精度.

结论:

  • 拟议的C-GBDL模型有效地为半监督的医疗图像细分生成高质量的伪标签.
  • 多尺度语义对应和双不确定性估计显著提高了细分性能.
  • 这种方法提供了一个有前途的解决方案,以降低注释成本并提高临床效用.