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

Blind Source Separation of Event-Related Potentials Using Recurrent Neural Network.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2025
Same author

Investigating the effects of focused attention (mantra) meditation on mismatch negativity: Insights into sensory and cognitive processing using an intensity oddball paradigm.

Neuroscience·2025
Same author

Localized estimation of event-related neural source activity from simultaneous MEG-EEG with a recurrent neural network.

Neural networks : the official journal of the International Neural Network Society·2024
Same author

Blind source separation of event-related potentials using a recurrent neural network.

bioRxiv : the preprint server for biology·2024
Same author

A simultaneous EEG-fNIRS dataset of the visual cognitive motivation study in healthy adults.

Data in brief·2024
Same author

Localized estimation of electromagnetic sources underlying event-related fields using recurrent neural networks.

Journal of neural engineering·2023

相关实验视频

Updated: Jun 21, 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

评估合成神经成像数据增强,以使用深度全卷积网络自动对脑瘤进行细分.

Fawad Asadi1, Thanate Angsuwatanakul1, Jamie A O'Reilly2

  • 1College of Biomedical Engineering, Rangsit University, Pathum Thani 12000, Thailand.

IBRO neuroscience reports
|July 15, 2024
PubMed
概括

使用StyleGAN2-ada的合成数据生成为自动化质瘤细分提供了最小的改进,这表明计算成本可能超过好处. 几何增强证明更有效地改善神经成像中的模型概括.

关键词:
脑组织细分 脑组织细分生成性的对抗性网络.质母细胞瘤 (glioblastoma) 是一个头部核磁共振扫描 (MRI) 是一个很好的方法.合成医学图像 合成医学图像

更多相关视频

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

8.9K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.7K

相关实验视频

Last Updated: Jun 21, 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
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

8.9K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.7K

科学领域:

  • 医疗成像医学成像
  • 人工智能的人工智能
  • 神经瘤学神经瘤学

背景情况:

  • 医学图像中的自动化质瘤细分对于治疗规划至关重要.
  • 深度学习模型需要广泛的,高质量的培训数据.
  • 现有的数据集可能不足以进行可靠的模型培训.

研究的目的:

  • 开发和评估用于增强数据的神经成像合成技术.
  • 用合成数据训练完全卷积网络 (U-nets) 来自动对质瘤进行细分.
  • 评估合成数据增强对细分性能和计算成本的影响.

主要方法:

  • 使用StyleGAN2-ada生成合成FLAIRMRI图像和相应的质瘤细分面具.
  • 合成数据被添加到一个真实数据集 (n=2751) 用于U-net培训.
  • 用和没有几何增强 (翻译,变焦,剪切) 训练U-net,并使用子系数进行评估.

主要成果:

  • 合成数据增强在子系数中产生了边际改善 (验证+0.0409,测试+0.0355).
  • 几何增强显著改善了模型的概括性,减少了跨数据集的性能变化.
  • 合成数据生成的计算费用是相当大的.

结论:

  • 合成数据增强为自动质瘤细分提供了适度的性能增长.
  • 几何增强对于改善U网的泛化更有好处.
  • 合成数据生成的当前计算成本很难为此应用程序辩护.