Jove
Visualize
联系我们

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

Spatiotemporal Encoding With Nonlinear Gradient Hardware Using Pulseq: From Principles to Practical Demonstration.

Magnetic resonance in medicine·2026
Same author

Quantification of valvular regurgitation by transthoracic 3D high pulse repetition frequency Doppler echocardiography.

European heart journal. Imaging methods and practice·2025
Same author

Deep learning intravoxel incoherent motion modeling: Exploring the impact of training features and learning strategies.

Magnetic resonance in medicine·2023
Same author

CT to MR registration of complex deformations in the knee joint through dual quaternion interpolation of rigid transforms.

Physics in medicine and biology·2021
Same author

Multiple-echo steady-state (MESS): Extending DESS for joint T<sub>2</sub> mapping and chemical-shift corrected water-fat separation.

Magnetic resonance in medicine·2021
Same author

Deep learning-based MR-to-CT synthesis: The influence of varying gradient echo-based MR images as input channels.

Magnetic resonance in medicine·2019
Same journal

Influence of gadolinium-based contrast agent (GBCA) on the diffusion weightings of breast lesions: an intra-patient analysis.

Magma (New York, N.Y.)·2026
Same journal

Evaluation of the diffusion time dependence of the IVIM effect based on realistic capillary flow simulations in mouse brain.

Magma (New York, N.Y.)·2026
Same journal

An evaluation of brain volume and cortical thickness measurement at 0.55 T.

Magma (New York, N.Y.)·2026
Same journal

Net zero emission MR imaging using a permanent 0.4 T magnet.

Magma (New York, N.Y.)·2026
Same journal

Special issue on "deuterium metabolic imaging".

Magma (New York, N.Y.)·2026
Same journal

Black-blood dynamic contrast-enhanced MRI of abdominal aortic aneurysms.

Magma (New York, N.Y.)·2026
查看所有相关文章
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

相关实验视频

Updated: Jun 14, 2025

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
12:49

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

12.7K

使用有限数据进行基于深度学习的图像重建:使用深度学习生成合成原始数据.

Frank Zijlstra1,2, Peter Thomas While3,4

  • 1Department of Radiology and Nuclear Medicine, St Olav's University Hospital, Postboks 3250 Torgarden, 7006, Trondheim, Norway. Frank.Zijlstra@stolav.no.

Magma (New York, N.Y.)
|August 29, 2024
PubMed
概括
此摘要是机器生成的。

合成数据生成提高了MRI重建质量,特别是当有限的原始数据可用时. 这种方法有效地利用现有的仅大小数据集来提高加速MRI扫描的性能.

关键词:
加速核磁共振成像 (MRI) 的使用.深度学习是一种深度学习.图像重建 图像重建综合数据 综合数据

更多相关视频

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects
10:16

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

12.2K
Cryo-Structured Illumination Microscopic Data Collection from Cryogenically Preserved Cells
11:55

Cryo-Structured Illumination Microscopic Data Collection from Cryogenically Preserved Cells

Published on: May 28, 2021

4.1K

相关实验视频

Last Updated: Jun 14, 2025

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
12:49

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

12.7K
Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects
10:16

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

12.2K
Cryo-Structured Illumination Microscopic Data Collection from Cryogenically Preserved Cells
11:55

Cryo-Structured Illumination Microscopic Data Collection from Cryogenically Preserved Cells

Published on: May 28, 2021

4.1K

科学领域:

  • 医疗成像医学成像
  • 人工智能的人工智能
  • 机器学习 机器学习

背景情况:

  • 深度学习擅长使用大量原始数据重建加速MRI扫描.
  • 原始MRI数据的有限可用性对训练深度学习模型构成了挑战.
  • 合成数据生成为增强小型数据集提供了一个潜在的解决方案.

研究的目的:

  • 调查合成数据生成的有效性,以改善加速MRI重建.
  • 评估合成数据对深度学习模型性能的影响,使用不同的数据集大小.
  • 探索仅用于大小数据集的合成原始数据创建的使用.

主要方法:

  • 一个敌对的自动编码器被用来从大小图像中生成合成相位和线圈灵敏度图.
  • 通过将生成的地图与大小图像相结合,创建了合成原始MRI数据.
  • 深度学习重建网络使用不同数量的真实和合成数据 (20-160次扫描) 进行训练,以实现四倍加速的MR任务.

主要成果:

  • 使用合成数据的训练减少了重建错误,特别是在较小的训练集中 (平均绝对错误降低了7.5%).
  • 对于较大的训练集,合成数据导致MAE略有增加 (高达2.6%).
  • 当使用合成数据来补充有限的真实原始数据时,观察到性能增长.

结论:

  • 合成原始数据生成有效地提高了MRI重建质量在数据有限的场景.
  • 这种方法可以利用更容易获得的仅大小数据集,克服原始MRI数据的稀缺性.
  • 合成数据为增强基于深度学习的加速MRI重建提供了有价值的策略.