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

Development of Eudragit L100/sodium alginate-carrageenan core/shell nanofiber patches for dermal delivery of omega-3 acid ethyl ester.

International journal of pharmaceutics·2026
Same author

A Roadmap to Perfused Skin: Defining the Next Generation of Research Questions in Cutaneous Tissue Engineering.

International journal of molecular sciences·2026
Same author

Association between model of anesthesiology care and outcomes after cataract surgery: a population-based cohort study.

CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne·2026
Same author

Stabilization of Niaouli essential oil-loaded gelatin nanofibers: characterization and in vitro evaluation.

Scientific reports·2026
Same author

Ibuprofen and nimesulide derivatives selectively induce apoptosis in HER2-positive breast cancer via inhibition of the PLA₂-COX-2-NF-κB pathway.

Molecular biology reports·2026
Same author

Synthesis, characterization, biological evaluation, and molecular modeling of novel nimesulide urea derivatives as potential MetAP2 inhibitors.

Scientific reports·2026

相关实验视频

Updated: Jul 20, 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

从小数据学习:通过深度学习从视网膜图像分类性别.

Aaron Berk1, Gulcenur Ozturan2, Parsa Delavari2

  • 1Department of Mathematics & Statistics, McGill University, Montréal, Canada.

PloS one
|August 3, 2023
PubMed
概括

深度学习模型可以使用小型数据集从视网膜底图像中对患者的性别进行分类. 这种方法克服了医学成像中的数据隐私挑战,以最小的数据实现了良好的性能.

更多相关视频

Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies
06:16

Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies

Published on: July 28, 2023

2.6K
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

1.5K

相关实验视频

Last Updated: Jul 20, 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
Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies
06:16

Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies

Published on: July 28, 2023

2.6K
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

1.5K

科学领域:

  • 眼科医生 眼科 眼科
  • 医疗成像医学成像
  • 计算机科学 计算机科学

背景情况:

  • 深度学习 (DL) 和卷积神经网络 (CNN) 显示出在医学成像中自动诊断的前景.
  • 视网膜底部成像适合自动分析,但通常需要大量的数据集.
  • 数据隐私和居住限制限制了临床环境中大量数据集的使用.

研究的目的:

  • 评估DL模型在小型数据集上的性能,以从视网膜底部图像中分类患者的性别.
  • 为应对医疗人工智能开发中的有限数据可用性的挑战.
  • 调查从 fundus 图像进行性别分类的可行性,这是以前未经定量化的特征.

主要方法:

  • 微调一个Resnet-152模型,修改了一个用于二进制分类的完全连接层.
  • 在使用私有 (DOVS) 和公共 (ODIR) 数据源的小型数据集上进行实验.
  • 用大约2500个基底图像评估模型性能.

主要成果:

  • 获得的测试AUC得分高达0.72 (95%CI:[0.67,0.77]) 在小数据集.
  • 与之前的工作相比,尽管数据集大小减少了1000倍,但仅表现出25%的性能下降.
  • 展示了成功的域调整,在一个数据集上训练的模型对另一个数据集进行了概括.
  • 强调了高质量的图像和组合的重要性,以最大限度地利用有限的数据来实现性能.

结论:

  • 二元分类,包括从视网膜图像进行性别分类,即使使用非常小的数据集也是可行的.
  • DL模型可以很好地对独立的数据源进行概括,克服分布变化.
  • 仔细的数据策划和组合对于在低数据场景中优化DL性能至关重要.
  • 这项研究验证了DL在严格的数据约束下用于医学成像应用的潜力.