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

HPV viral load predicts immune exhaustion and prognosis in cervical neoplasia.

Frontiers in immunology·2026
Same author

Parental age and age gap in relation to neonatal health: a cohort study in eastern China.

Frontiers in reproductive health·2026
Same author

Decoding the Malignant Potential of Hydatidiform Moles through Fibroblast and LAIR2 Signatures.

Cancer communications (London, England)·2026
Same author

Correction: EZH2 blockade reverses doxorubicin resistance by inducing metabolic vulnerability and enhancing DNA damage in breast cancer.

Frontiers in pharmacology·2026
Same author

Magnetic resonance imaging radiomic phenotypes stratify response and define immune states in very high-risk, nonmuscle-invasive bladder cancer treated with tislelizumab plus nanoparticle albumin-bound paclitaxel.

Cancer·2026
Same author

Analysis of SOX2 and SOX17 expression in cervical HPV-associated adenocarcinoma in situ: Correlation with histologic variants.

Annals of diagnostic pathology·2026

相关实验视频

Updated: Jun 13, 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

一个细分任何东西以模型为导向和基于匹配的半监督细分框架用于医学成像.

Guoping Xu1, Xiaoxue Qian1, Hua-Chieh Shao1

  • 1The Medical Artificial Intelligence and Automation (MAIA) Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA.

Medical physics
|March 29, 2025
PubMed
概括

SAM-Match通过使用Segment Anything Model (SAM) 来提高基于匹配的框架的伪标签质量,提高半监督医疗图像细分,在有限的数据中实现高精度.

关键词:
基于匹配的框架.细分任何东西模型模型.半监督的细分化方式

更多相关视频

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

982
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.5K

相关实验视频

Last Updated: Jun 13, 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: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

982
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.5K

科学领域:

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

背景情况:

  • 半监督细分利用有限的标记数据与未标记数据一起用于改进模型训练.
  • 基于匹配的框架依赖于一致性约束,但与低质量的伪标签作斗争.
  • 细分任何模型 (SAM) 由于广泛的预培训,提供了强大的概括性.

研究的目的:

  • 将分段任何模型 (SAM) 集成到基于匹配的框架中,以增强半监督的医疗图像细分.
  • 使用SAM的功能,提高半监督学习中生成的伪标签的质量.
  • 利用SAM的泛化能力,在各种医学成像领域实现更好的性能.

主要方法:

  • 提出SAM-Match,这是一个结合SAM和基于匹配的方法的新框架,用于半监督的医疗图像细分.
  • 使用预训练的基于匹配的模型来生成高可靠性预测,以便快速创建.
  • 采用精心调整的基于SAM的方法,生成提示和未标记的数据,以产生高质量的伪标签用于培训.

主要成果:

  • 在心脏MRI (ACDC),乳房超声波 (BUSI) 和肝脏MRI数据集上,SAM-Match实现了强大的细分性能.
  • 具有最小标签数据的子得分很高:ACDC (3个标签) 的89.36%,BUSI (30个标签) 的59.35%,MRLiver (3个标签) 的80.04%.
  • 统计学意义证实了通过Wilcoxon签名等级测试对现有方法的改进.

结论:

  • SAM-Match框架有效地解决了针对SAM的快速生成和基于匹配的模型的伪标签方面的挑战.
  • 显示出加速在医学成像中采用半监督学习的巨大潜力,特别是在数据稀缺的情况下.
  • 代码和数据将公开发布,以促进进一步的研究和应用.