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

相关实验视频

Updated: Jun 23, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

386

MultiTrans:用于医疗图像细分的多分支变压器网络.

Yanhua Zhang1, Gabriella Balestra2, Ke Zhang3

  • 1Department of Electronics and Telecommunications, Politecnico di Torino, Corso Duca degli Abruzzi 24, Turin, 10129, Italy; School of Astronautics, Northwestern Polytechnical University, 127 West Youyi Road, Xi'an, 710072, China.

Computer methods and programs in biomedicine
|June 15, 2024
PubMed
概括

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

Integrating Radiomics and Machine Learning to Improve Fluorescence Image Segmentation in in vitro models.

Studies in health technology and informatics·2026
Same author

Data Warehouse to Support Clinical Trials in Dentistry.

Studies in health technology and informatics·2026
Same author

The challenges of national health data ecosystems in feeding the European health data space: the Italian example.

Frontiers in medicine·2025
Same author

Impact of the auscultation area on heart sound waveforms for the estimation of Cardiac Time Intervals.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Pilot study on the separability of the native heart sounds and device support noise in patients implanted with left ventricular assist devices.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Intelligent System for Automated Spheroid Segmentation Using Machine Learning.

Studies in health technology and informatics·2025

本研究介绍了MultiTrans,这是一个高效的变压器架构,用于医疗图像细分. 它通过有效处理高分辨率特征并聚合多个规模的全球和本地信息来实现卓越的准确性.

科学领域:

  • 医学图像分析 医学图像分析
  • 深度学习 (Deep Learning) 是一种深度学习.
  • 计算机视觉 计算机视觉

背景情况:

  • 卷积神经网络 (CNN) 主导医疗图像细分,但与全球背景作斗争.
  • 变压器提供全球上下文,但在计算上是低效的,通常需要降低采样或基于补丁的处理.
  • 补丁智能操作限制了变压器在捕获细粒度,像素级细节方面,这对于细分至关重要.

研究的目的:

  • 为医疗图像细分中的变压器开发一种内存和计算效率高的自我注意模块.
  • 设计一种新的多分支变压器 (MultiTrans) 架构,能够处理多尺度的特征.
  • 改善医疗图像中全球背景和精细空间细节的提取.

主要方法:

  • 提出了一个有效的自我注意 (ESA) 模块,以使得对高分辨率特征的推理.
  • 引入了多分支变压器 (MultiTrans) 架构,在不同的CNN级别上有四个并行变压器分支.
  • 通过混合CNN-变压器方法聚合多个规模的全球背景和多个规模的本地特征.

主要成果:

  • 在三种不同的医疗图像数据集 (Synapse,ACDC,M&Ms) 中,MultiTrans实现了最高的细分精度.
  • 与标准自我注意力 (SSA) 相比,提出的高效自我注意力 (ESA) 显著降低了训练记忆 (18.77%) 和计算复杂性 (20.68% FLOP).
关键词:
腹部多器官细分 腹部多器官细分心脏细分是指心脏的细分.深度学习是一种深度学习.有效的自我注意力.医疗图像细分 医疗图像细分多分支变压器多分支变压器

更多相关视频

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.3K
A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
09:41

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery

Published on: May 20, 2016

12.3K

相关实验视频

Last Updated: Jun 23, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

386
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.3K
A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
09:41

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery

Published on: May 20, 2016

12.3K
  • 欧元局保持或略提高了细分精度,同时大幅减少了模型参数 (74.07%的减少).

    • 结论:

    • 多传输网络在医疗图像细分任务上表现出强大且可通用的性能.
    • 废弃性研究证实了提议的高效自我照顾 (ESA) 模块的效率和有效性.
    • 开发的架构解决了现有方法的局限性,为医疗图像细分提供了更有效,更准确的解决方案.