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

A near-infrared fluorescent probe for facile fluoride detection in environmental and biological systems.

Chemical communications (Cambridge, England)·2026
Same author

Biolipid Film-Fused Electrochemiluminescence for Multipurpose In Situ Bioassays.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same author

Dual-key cooperatively activated DNA regulator for controlling mitochondria-lysosome interactions.

Nature communications·2025
Same author

Prussian Blue Nanozyme-Based Dry-Dipstick Reflectometry for Multiplex Detection of Low Molecular Weight Metabolites.

Analytical chemistry·2025
Same author

Single-Protein Determinations by Magnetofluorescent Qubit Imaging with Artificial-Intelligence Augmentation at the Point-Of-Care.

ACS nano·2025
Same author

Controllable mitochondrial regulation based on photo-triggered DNA circuitry.

Journal of materials chemistry. B·2025

相关实验视频

Updated: Sep 9, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

491

通过并行多尺度变压器-CNN聚合网络进行高效的3D生物医学图像细分

Wei Liu1, Yuxiao He1, Tiantian Man1

  • 1School of Mechanical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China.

Chemical & biomedical imaging
|August 29, 2025
PubMed
概括

一个新的混合并行变压器 (MPSTrans) 模型通过有效捕捉全球和本地特征来增强3D生物医学图像细分. 这种先进的深度学习方法提高了临床诊断和手术规划的准确性和效率.

关键词:
3D生物医学图像细分卷积神经网络多级特征提取平行架构连接变压器

更多相关视频

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

2.9K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.3K

相关实验视频

Last Updated: Sep 9, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

491
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

2.9K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.3K

科学领域:

  • 医学成像分析
  • 医疗保健中的人工智能
  • 用于细分的深度学习

背景情况:

  • 准确的3D生物医学图像细分对于临床诊断,手术和预后至关重要.
  • 现有的深度学习模型在同时捕获全球和本地图像特征以进行精确的细分方面面临挑战.

研究的目的:

  • 为改进3D生物医学图像分析开发先进的分段解决方案 - - 混合并行变频器 (MPSTrans).
  • 解决当前方法在综合特征捕获和多尺度同步方面的局限性.

主要方法:

  • 介绍了MPSTrans,一个新的深度学习模型,在U型框架内使用3D-MPST块.
  • 在解码器中实现了深度监控以实现层次表示学习.
  • 评估结肠癌,多器官和多模式数据集的性能.

主要成果:

  • 在结肠癌数据中,MPSTrans显著改善了子相似系数 (DSC) 并减少了95%的豪斯多夫距离 (HD95).
  • 实现了56.7%的计算负载 (GFLOPs) 的减少.
  • 在公共数据集 (MSD,BCV,ACD) 上表现优于Swin UNETR,UNETR,nnU-Net,PHTrans等主流方法.

结论:

  • MPSTrans为3D生物医学图像细分提供了强大而可适应的解决方案,提高了诊断能力.
  • 该模型能够捕捉全面的特征并减少计算负载,使其成为医学成像分析的最先进工具.