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

Understanding older adults' perception, acceptance, and adoption of smart home technologies.

PloS one·2026
Same author

BioFuse: an embedding fusion framework for biomedical foundation models.

PloS one·2026
Same author

Exploring techniques to distinguish between real images and those generated using stable diffusion XL.

PloS one·2026
Same author

Endometrial whole-slide images dataset for detection of malignancy in endometrial biopsies.

GigaScience·2025
Same author

Cervical whole-slide images dataset for multiclass classification.

GigaScience·2025
Same author

SurGen: 1020 H&E-stained whole-slide images with survival and genetic markers.

GigaScience·2025

相关实验视频

Updated: Jul 8, 2025

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.2K

超大规模多式联络成像数据集的分类

Craig Macfadyen1, Ajay Duraiswamy1, David Harris-Birtill1

  • 1University of St Andrews, St Andrews, United Kingdom.

PLOS digital health
|December 13, 2023
PubMed
概括
此摘要是机器生成的。

深度学习模型可以根据模式 (CT,MRI,PET,X射线) 准确地分类数以百万计的医疗图像. 这加快了诊断成像数据的检索速度,以96%的准确度改善了临床结果.

更多相关视频

Multimodal Hierarchical Imaging of Serial Sections for Finding Specific Cellular Targets within Large Volumes
11:19

Multimodal Hierarchical Imaging of Serial Sections for Finding Specific Cellular Targets within Large Volumes

Published on: March 20, 2018

10.5K
Multimodal Optical Imaging Platform for Studying Cellular Metabolism
04:47

Multimodal Optical Imaging Platform for Studying Cellular Metabolism

Published on: June 6, 2025

332

相关实验视频

Last Updated: Jul 8, 2025

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.2K
Multimodal Hierarchical Imaging of Serial Sections for Finding Specific Cellular Targets within Large Volumes
11:19

Multimodal Hierarchical Imaging of Serial Sections for Finding Specific Cellular Targets within Large Volumes

Published on: March 20, 2018

10.5K
Multimodal Optical Imaging Platform for Studying Cellular Metabolism
04:47

Multimodal Optical Imaging Platform for Studying Cellular Metabolism

Published on: June 6, 2025

332

科学领域:

  • 人工智能的人工智能
  • 医疗成像医学成像
  • 计算机科学 计算机科学

背景情况:

  • 超大规模的多模式数据集,包含数百万张图像,需要有效的分类方法用于诊断成像数据检索.
  • 准确的医疗图像分类模式对于加速临床结果和研究至关重要.

研究的目的:

  • 为了证明在超大规模,多模式数据集上训练的深度神经网络的有效性,以准确的医学图像模式分类.
  • 评估不同卷积神经网络 (CNN) 架构 (ResNet-50,ResNet-18,VGG16) 在分类医学成像模式中的性能.

主要方法:

  • 通过结合102个不同的数据集,创建了一个包含450万张异质医疗图像的数据集.
  • 训练ResNet-50,ResNet-18和VGG16模型将图像分为四种模式:计算机断层扫描 (CT),磁共振成像 (MRI),正子发射断层扫描 (PET) 和X射线.
  • 模型的性能被评估在未见的数据,测量分类准确度和平衡准确度.

主要成果:

  • 性能最好的模型在未见数据上实现了96%的分类准确性,与EfficientNets和Vision Transformers (ViTs) 等更复杂的模型相比或超过.
  • 顶级模型实现了86%的平衡精度.
  • 该研究证实了在超大规模的多式联络数据集上训练深度学习CNN的可行性.

结论:

  • 在超大规模多式联络数据集上训练的深度学习模型可以在分类医疗成像方式方面实现高准确性.
  • 这些模型对于涉及大规模医疗图像存储库和国家医疗机构的现实应用具有重大潜力.
  • 未来的研究可以将这种分类能力扩展到包括3D扫描.