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

Bioactive compounds from <i>Ficus hirta</i> Vahl extract: Molecular mechanisms underlying their anti-inflammatory and antibacterial activities.

Biochemistry and biophysics reports·2026
Same author

Erratum for Zeng et al., "Predictive value of hepatic, hematological, and immunological markers and their temporal dynamics in chronic hepatitis B functional cure".

Microbiology spectrum·2026
Same author

Fluorescent labeled enzyme immobilized on chitosan-coated magnetic microspheres for potential cyclooxygenase-2 inhibitors screening accompanied with molecular modeling and in situ cell imaging.

Mikrochimica acta·2026
Same author

Penicinines A and B: two new 2-pyridone derivatives from a marine-derived fungus <i>Penicillium</i> sp. BTBU20218001.

Natural product research·2026
Same author

Host-viral interaction of HBV infection revealed by single-cell transcriptome jointly profiling the viral replication state.

Hepatology (Baltimore, Md.)·2026
Same author

Berberine Suppresses Pathogenic Fungus <i>Aspergillus fumigatus</i> Hyphal Growth via Mitochondrial Fragmentation-Induced ROS Elevation and Hog1-MAPK Activation.

ACS infectious diseases·2025

相关实验视频

Updated: Jul 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

2.8K

基于多尺度特征和风格转移的视网膜血管细分.

Caixia Zheng1,2, Huican Li2, Yingying Ge1

  • 1Jilin Animation Institute, Changchun 130013, China.

Mathematical biosciences and engineering : MBE
|February 2, 2024
PubMed
概括

这项研究引入了一种新的深度学习网络 (MSFST-NET),用于改进视网膜血管细分,提高跨领域数据和没有复杂模型的小血管的性能.

关键词:
深度学习是一种深度学习.多个尺度的特征.伪标签学习学习视网膜血管细分器的细分风格转移 风格转移 风格转移

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

405
Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography
07:23

Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography

Published on: March 26, 2020

7.6K

相关实验视频

Last Updated: Jul 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

2.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

405
Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography
07:23

Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography

Published on: March 26, 2020

7.6K

科学领域:

  • 眼科医生 眼科 眼科
  • 医疗成像医学成像
  • 计算机视觉 计算机视觉

背景情况:

  • 视网膜血管细分对于诊断眼睛疾病至关重要.
  • 当前的深度学习方法在跨领域数据和小容器细分方面扎.
  • 过于复杂的模型在实际应用中存在挑战.

研究的目的:

  • 为改善视网膜血管细分提出一种新的轻量级网络 (MSFST-NET).
  • 增强模型处理跨领域数据集和细分小血管的能力.
  • 为了避免过于复杂的模型架构.

主要方法:

  • 开发了一个具有选择性内核 (SK) 模块的MSF-Net,用于多规模的特征提取.
  • 引入了风格传输模块,以减少域差异.
  • 实施伪标签学习策略以促进概括.

主要成果:

  • 在MSFST-NET中,小血管的细分有所改善.
  • 风格转移和伪标签有效地改善了跨域性能.
  • 在DRIVE和CHASE_DB1数据集的实验中,与最先进的方法相比,显示出优异的结果.

结论:

  • MSFST-NET为视网膜血管细分提供了有效的解决方案,特别是在跨领域的挑战中.
  • 提出的方法可以提高模型的概括性和细分精度.
  • 这种方法为临床应用提供了更强大,更有效的工具.