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

CrossModalSync: joint temporal-spatial fusion for semantic scene segmentation in large-scale scenes.

Scientific reports·2025
Same author

The relationship between sleep duration and obesity among risky drinking workers: using the Korea National Health and Nutrition Survey.

Korean journal of family medicine·2025
Same author

Enhancing generalization of medical image segmentation via game theory-based domain selection.

Journal of biomedical informatics·2025
Same author

Learning generalizable visual representation via adaptive spectral random convolution for medical image segmentation.

Computers in biology and medicine·2024
Same author

A Study on the Job Characteristics of Police Officers and the Status of Musculoskeletal Disorders.

Healthcare (Basel, Switzerland)·2024
Same author

Volumetric Imitation Generative Adversarial Networks for Anatomical Human Body Modeling.

Bioengineering (Basel, Switzerland)·2024

相关实验视频

Updated: Jul 2, 2025

Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing
11:36

Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing

Published on: February 9, 2022

2.8K

3D-DGGAN:一个数据引导的生成对抗网络,用于医疗图像生成的高保真性.

Jion Kim, Yan Li, Byeong-Seok Shin

    IEEE journal of biomedical and health informatics
    |February 28, 2024
    PubMed
    概括

    本研究介绍了一种新的数据引导生成对抗网络,用于创建高保真度3D医疗图像. 该方法有效地生成现实的3D图像,即使训练数据有限,克服了以前的局限性.

    科学领域:

    • 医疗成像医学成像
    • 人工智能的人工智能
    • 计算机视觉 计算机视觉

    背景情况:

    • 3D医学图像的有限可用性阻碍了分类,细分和检测方面的研究.
    • 现有的二维生成方法与三维解剖结构作斗争,导致切片不连续性.
    • 当前的3D生成网络需要大量的数据,这往往是不可用的,导致不充分的培训和低保真度输出.

    研究的目的:

    • 提出一种新的数据导向生成对抗网络 (GAN),用于高准确度的3D医学图像生成.
    • 为了应对从有限的数据集中生成现实的3D医学图像的挑战.
    • 提高人工智能生成的医学成像数据的准确性和细节性.

    主要方法:

    • 数据引导的GAN使用生成器从从真实数据中提取的参考代码创建图像.
    • 该生成器产生噪音和无噪音的解码图像,以评估与真实图像对比的参考代码忠实性.
    • 使用多元分辨器 (体积,板块和切片) 通过分析不同颗粒度的图像来提高保真度.

    主要成果:

    • 提出的方法成功地使用少量真实训练数据生成高准确度的3D医疗图像.
    • 使用Fréchet初始距离 (FID) 和最大平均差异 (MMD) 的定量比较显示出比现有方法更高的性能.

    更多相关视频

    3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects
    08:15

    3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects

    Published on: August 4, 2020

    6.4K
    3D Printing of Preclinical X-ray Computed Tomographic Data Sets
    11:06

    3D Printing of Preclinical X-ray Computed Tomographic Data Sets

    Published on: March 22, 2013

    40.5K

    相关实验视频

    Last Updated: Jul 2, 2025

    Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing
    11:36

    Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing

    Published on: February 9, 2022

    2.8K
    3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects
    08:15

    3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects

    Published on: August 4, 2020

    6.4K
    3D Printing of Preclinical X-ray Computed Tomographic Data Sets
    11:06

    3D Printing of Preclinical X-ray Computed Tomographic Data Sets

    Published on: March 22, 2013

    40.5K
  • 多元组件区分器有效地区分了真实与生成的图像,提高了整体现实性.

    • 结论:

    • 数据引导的GAN为生成高保真度3D医疗图像提供了强大的解决方案,特别是在数据稀缺的场景中.
    • 提出的方法克服了以前2D和3D生成方法的局限性.
    • 这一进步对医学成像研究产生了重大影响,使得模型培训和开发更加稳健.