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

Metagenome-assembled genome of <i>Oscillospiraceae</i> bacterium strain ZGZL, an anaerobic chloromethane-degrading bacterium enriched from rice paddy soil.

Microbiology resource announcements·2026
Same author

Transcriptomic analysis of response to high-temperature stress in cotton.

Plant physiology and biochemistry : PPB·2026
Same author

Hepatoprotective Mechanisms of Lactiplantibacillus plantarum SCS7 Cell-Free Extract Against Aflatoxin B1 Toxicity.

Probiotics and antimicrobial proteins·2026
Same author

Metabolic engineering of Halomonas cupida for co-mineralization of phenol and p-nitrophenol in high-saline wastewater.

Journal of environmental management·2026
Same author

ITS-ShipFormer: An Informative Token Selection Former for SAR Ship Recognition.

IEEE transactions on neural networks and learning systems·2026
Same author

Evaluation of Three Treatments for the Resource Utilization of Cephalosporin C Fermentation Residue.

Toxics·2026

相关实验视频

Updated: Jul 8, 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

2.8K

使用三路Unet模型进行视网膜图像细分.

Ruihua Liu1, Wei Pu2,3, Haoyu Nan4,5

  • 1School of Artificial Intelligence, Chongqing University of Technology, Chongqing, China. lruih@cqut.edu.cn.

Scientific reports
|December 19, 2023
PubMed
概括
此摘要是机器生成的。

这项研究引入了一种新的三路Unet模型,用于精确的视网膜图像细分,提高检测血管的准确性. 改进后的模型显著优于现有方法,有助于医学诊断.

更多相关视频

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

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.8K
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

2.5K

相关实验视频

Last Updated: Jul 8, 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

2.8K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.8K
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

2.5K

科学领域:

  • 医疗成像医学成像
  • 计算机视觉 计算机视觉
  • 生物医学工程 生物医学工程

背景情况:

  • 无监督的图像细分提供了灵活性,但由于数据的变化和噪声,在视网膜成像方面面临着挑战.
  • 视网膜图像细分的手动调整可能会导致不准确的结果,比如血管的过度或不足细分.

研究的目的:

  • 开发一种新的监督细分网络,以提高视网膜图像细分的精度.
  • 为了提高视网膜血管细分的准确性和可靠性,以提供诊断辅助.

主要方法:

  • 开发了一种三路 Unet 模型 (TP-Unet),将 Haar 波量变换集成到 Unet 和 Seg.Net 中,用于高频信息提取 (HaarNet).
  • 通过将自动编码 (AE) 和深度监督学习 (DSL) 整合在一起,TP-Unet模型进一步改进为TP-Unet+AE+DSL.
  • 在公共DRIVE和CHASE视网膜图像数据集上进行了实验.

主要成果:

  • 拟议的TP-Unet+AE+DSL模型在DRIVE数据集上实现了0.8291的Dice系数和0.8184的灵敏度,超过了UNET模型.
  • 在CHASE数据集上,该模型实现了0.8162的子系数,0.8242的灵敏度和0.9664的准确性,超过了UNET模型.
  • 结果显示,视网膜血管细分的准确性和可靠性得到了显著改善.

结论:

  • 新型TP-Unet+AE+DSL模型在视网膜血管细分方面提供了优越的性能,与标准Unet.
  • 拟议的方法提高了精度和可靠性,显示了临床诊断支持的潜力.
  • 准确的视网膜血管细分对于早期检测和治疗各种眼睛疾病至关重要.