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

Distinct medication-state modulation of motor-cortical low-beta power in tremor-dominant and postural instability/gait difficulty Parkinson's disease: a source-space resting-state EEG study.

Frontiers in neurology·2026
Same author

All-two-dimensional, ion-gating synaptic transistors for high-temperature and ultralow-energy-consumption neuromorphic applications.

Science advances·2026
Same author

A grading system based on multiparametric MRI-predicted tumor perfusion, invasiveness and texture (GOLDS-PIT) for surgical difficulty assessment in pituitary macroadenomas: A prospective single-centre cohort study.

World neurosurgery·2026
Same author

Disruption of the <i>UPC2</i> Gene Enhances Fluconazole Antifungal Activity by Inhibiting <i>HAC1</i> mRNA Splicing in <i>Candida albicans</i>.

Pathogens (Basel, Switzerland)·2026
Same author

Dual role and metabolic reprogramming: Key amino acids in marinade bidirectionally control aflatoxin B<sub>1</sub> biosynthesis in Aspergillus parasiticus.

Food microbiology·2026
Same author

Overexpression of FOXO1 may suppress proliferation and migration and correlate with tumor immune cells in nasopharyngeal carcinoma.

Translational cancer research·2026

相关实验视频

Updated: Jun 18, 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

基于角色交换的自我培训半监督框架,用于复杂的医疗图像细分.

Yonghuang Wu, Guoqing Wu, Jixian Lin

    IEEE transactions on neural networks and learning systems
    |August 2, 2024
    PubMed
    概括

    这项研究引入了一种新的半监督模型来分割复杂的医疗图像,显著减少注释需求. 双向自我训练方法在最小的标记数据下实现了高准确度.

    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:

    • 精确细分复杂的医学图像,如血管和肺网络是具有挑战性的,因为需要许多微小的目标注释.
    • 完全监督的深度学习模型需要大量的手动注释,阻碍了复杂结构的细分.

    研究的目的:

    • 为复杂的医疗图像细分开发一个高效的半监督模型.
    • 减少对大型注释数据集的依赖,用于医疗图像细分任务.

    主要方法:

    • 提出了一种双向的自我训练范式,基于模型可靠性,动态交换教师-学生角色.
    • 引入了非对称监督 (AS) 和层次双学生 (HDS) 结构,以防止在小数据集上模型崩.
    • 实现双向蒸损失与角色交换 (RE) 和全球-本地意识的一致性损失,以获得稳定的特征匹配.

    主要成果:

    • 拟议的半监督模型在公共和私人数据集上显著优于现有的方法.
    • 实现了与完全监督的方法可比的性能,仅占标签成本的5%.
    • 证明了稳定的相互促进和全球和地方特征的有效匹配.

    结论:

    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

    381
    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 18, 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
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    381
    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
  • 双向自我训练模型为复杂的医疗图像细分提供了高效的解决方案,要求最小的注释.
  • 该AS战略和HDS结构成功地解决了对小规模注释数据的培训的挑战.
  • 这种方法代表了有效和准确的医疗图像细分的重大进步.