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

In Vivo Phenotyping of Dopaminergic Neurodegeneration in Zebrafish Larvae Using Behavioral Analysis and High-Content Imaging.

Journal of visualized experiments : JoVE·2026
Same author

Preclinical evaluation of [<sup>68</sup>Ga]Ga-MRCP: a potential radiotracer for imaging PD-L1 expression in colorectal cancer.

Bioorganic chemistry·2026
Same author

When One Sequence Is Enough-And When It Isn't.

Radiology. Artificial intelligence·2026
Same author

Transplant-ready? Evaluating AI lung segmentation models in candidates with severe lung disease.

Journal of medical imaging (Bellingham, Wash.)·2026
Same author

Beyond technical access in digital eldercare: how ethical lag shapes stratified responsiveness to institutional welfare in rural China.

BMC geriatrics·2026
Same author

Modulation of lysozyme amyloid fibrillation by three structurally similar dihydroxy-coumarins: A comparative analysis.

Bioorganic chemistry·2026

相关实验视频

Updated: Jul 19, 2025

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.1K

医学图像分析的任何细分模型:一项实验性研究.

Maciej A Mazurowski1, Haoyu Dong2, Hanxue Gu2

  • 1Department of Radiology, Duke University, Durham, NC, 27708, USA; Department of Electrical and Computer Engineering, Duke University, Durham, NC, 27708, USA; Department of Computer Science, Duke University, Durham, NC, 27708, USA; Department of Biostatistics & Bioinformatics, Duke University, Durham, NC, 27708, USA.

Medical image analysis
|August 18, 2023
PubMed
概括

细分任何模型 (SAM) 在医疗图像细分方面表现多样,在某些数据集上表现出色,但在其他数据集上扎. 与其他方法相比,代提示为SAM提供了有限的收益.

关键词:
深度学习是一种深度学习.基金会模型 基金会模型分段化 分段化 分段化 分段化

更多相关视频

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.6K
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

相关实验视频

Last Updated: Jul 19, 2025

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.1K
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.6K
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

科学领域:

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

背景情况:

  • 医疗图像细分至关重要,但由于有限的注释数据而受到阻碍.
  • 像分段任何模型 (SAM) 这样的基础模型提供了潜在的解决方案.
  • 在自然图像上训练的SAM需要对专门的医疗领域进行评估.

研究的目的:

  • 为了广泛评估SAM的零射击医疗图像细分能力.
  • 将SAM的性能与其他交互式细分方法进行比较.
  • 确定影响SAM在各种医疗数据集中的细分精度的因素.

主要方法:

  • 评估了19个不同的医学成像数据集 (各种模式和解剖学) 的SAM.
  • 模拟使用点和框提示符的交互式细分.
  • 将SAM的性能 (IoU) 与RITM,SimpleClick和FocalClick进行了比较.

主要成果:

  • 在不同数据集中,SAM的性能差异很大 (脊柱MRI的IoU为0.1135,部X射线的IoU为0.8650).
  • 框提示的性能优于点提示的性能;对于有很好的界限的对象,性能更好.
  • SAM超越了使用单点提示的其他方法,但代提示带来了有限的改进.

结论:

  • 对于特定的医学成像任务,SAM展示了令人印象深刻的零射击细分.
  • 其他医疗数据集的性能中等到差,需要仔细应用.
  • SAM有可能实现自动化医疗图像细分,但域特定验证至关重要.