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

One-step microwave-assisted synthesis of nitrogen-doped carbon quantum dots for highly sensitive and selective fluorescence detection of mercury (II) ion in food.

Food chemistry: X·2026
Same author

Comparative effects of four back squat prescription models on explosive performance and sprint outcomes in resistance-trained men.

Frontiers in physiology·2026
Same author

Design and application of a FA-based molecularly imprinted sensor for screening anti-Fusariumoxysporum substances from Lanzhou lily endophytes.

Bioelectrochemistry (Amsterdam, Netherlands)·2026
Same author

Nomogram prediction model for pain recurrence in patients with trigeminal neuralgia after microvascular decompression.

Neurosurgical review·2026
Same author

Safety and efficacy of decompressive craniectomy versus standard craniotomy for large acute epidural haematoma with tentorial herniation in China (PREDICT-AEDH): a nationwide, multicentre, open-label, parallel-group, randomised controlled trial.

The Lancet. Neurology·2026
Same author

Electronic modulation of Fe-N<sub>4</sub> sites by second-shell sulfur for enhanced oxygen reduction electrocatalysis.

Chemical communications (Cambridge, England)·2026

相关实验视频

Updated: Jun 4, 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

一个改进的多尺度特征提取网络用于医疗图像细分.

Haoyu Guo1,2, Liuliu Shi1,2,3, Jinlong Liu4,5,6

  • 1School of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai, China.

Quantitative imaging in medicine and surgery
|December 19, 2024
PubMed
概括

这项研究引入了一个新的深度学习模型Res2-CD-UNet,用于增强医疗图像细分. 该模型提高了组织与背景分离的准确性,优于现有方法.

关键词:
多尺度的特点是多尺度的特点.道融合功能是一个道融合功能.全球都在关注这个问题.细分化 细分化的细分化空间信息就是空间信息.

更多相关视频

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

367
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.4K

相关实验视频

Last Updated: Jun 4, 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
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

367
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.4K

科学领域:

  • 医疗成像医学成像
  • 计算机视觉 计算机视觉
  • 深度学习 (Deep Learning) 是一种深度学习.

背景情况:

  • U-Net模型推进了医疗图像细分,但与空间信息丢失和背景噪声作斗争.
  • 现有的方法在医学图像中面临规模变化和复杂组织结构的挑战.
  • 准确的细分对于诊断和治疗规划至关重要.

研究的目的:

  • 开发一个先进的深度学习模型,以改善医疗图像细分.
  • 解决现有U-Net架构的局限性,特别是空间信息丢失和背景噪声.
  • 为了增强多尺度特征的提取,以实现更精确的组织分离.

主要方法:

  • 开发了Res2Net-ConvFormer-Dilation-UNet (Res2-CD-UNet),这是一个新的U形网络.
  • 集成的Res2Net作为强大的特征提取的支柱.
  • 包含一个卷积式变压器,以提高全球关注度,以及一个通道功能融合块 (CFFB) 来减轻背景噪声.

主要成果:

  • 在Synapse数据集上获得了83.92%的平均子相似系数 (DSC),超过了1.96%的次优模型.
  • 在Seg.A.2023数据集上获得了93.27%的平均DSC,证明了卓越的性能.
  • 显示了多个器官的细分精度的提高,在Synapse数据集的八个中,有四个器官的最佳结果.

结论:

  • 该Res2-CD-UNet模型显著提高了医疗图像细分的准确性.
  • 该网络有效地提取多个尺度的特征,并最大限度地降低背景噪声.
  • 这种深度学习方法为复杂的医学图像分析挑战提供了有希望的解决方案.