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

DRCmpVis: Visual Comparison of Physical Targets in Mobile Diminished and Mixed Reality.

IEEE transactions on visualization and computer graphics·2024
Same author

Prevalence and Rates of New Diagnosis and Missed Diagnosis of Diabetes Mellitus among 35-74-year-old Residents in Urban Communities in Southwest China.

Biomedical and environmental sciences : BES·2019
Same author

Insights into the degradation and detoxication mechanisms of aqueous capecitabine in electrochemical oxidation process.

Chemosphere·2019
Same author

Resolving electrical stimulus triggered molecular binding and force modulation upon thrombin-aptamer biointerface.

Journal of colloid and interface science·2019
Same author

Effect of periodic temperature shock on nitrogen removal performance and microbial community structure in plug-flow microaerobic sludge blanket.

Chemosphere·2019
Same author

Central Nervous System Complications after Allogeneic Hematopoietic Stem Cell Transplantation in Children.

Acta haematologica·2019

相关实验视频

Updated: Jun 9, 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

稀疏编码启发了LSTM和自我注意力集成,用于医疗图像细分.

Zexuan Ji, Shunlong Ye, Xiao Ma

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
    |October 22, 2024
    PubMed
    概括

    这项研究将长期短期记忆 (LSTM) 与自我注意 (SA) 稀疏编码相结合,以改善医疗图像细分. 这种新的方法增强了神经网络.

    科学领域:

    • 医疗成像医学成像
    • 人工智能的人工智能
    • 计算机视觉 计算机视觉

    背景情况:

    • 准确的医学图像细分对于临床诊断和分析至关重要.
    • 整合上下文关系可以增强神经网络的表达能力.
    • 长期短期记忆 (LSTM) 和自我注意 (SA) 捕捉全球依赖性,但通常是单独的模块.

    研究的目的:

    • 展示LSTM设计与SA稀疏编码的创新整合.
    • 为了利用LSTM的状态压缩和历史数据保留,为SA.
    • 通过整合时间信息来增强SA的稀疏编码和全球依赖性捕获.

    主要方法:

    • 开发了一种新的方法,使用LSTM状态的线性组合用于SA的查询,键和值 (QKV) 矩阵.
    • 引入了两个模块,将SA矩阵集成到LSTM状态设计中.
    • 将这些模块嵌入到U形卷积神经网络架构中,用于2D和3D医学图像.

    主要成果:

    • 拟议的模块显著提高了四个数据集的医疗图像细分任务的性能.
    • 超过了各种基线,包括那些已经使用上下文模块的基线.
    • 在下游细分任务中提高预测准确性.

    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

    370
    Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
    06:48

    Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

    Published on: January 7, 2019

    8.8K

    相关实验视频

    Last Updated: Jun 9, 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

    370
    Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
    06:48

    Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

    Published on: January 7, 2019

    8.8K

    结论:

    • 集成的LSTM和SA稀疏编码方法有效地模拟全球依赖关系.
    • 这些新型模块可以提高医疗图像细分的准确性,而无需额外的计算成本.
    • 这种整合为推进医学图像分析提供了一个有希望的方向.