Search research articles

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 Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

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

AI weather prediction toward multi-scale earth system prediction: status and pathways.

Science bulletin·2026

Same author

Plant-derived extracellular vesicles as emerging cardioprotective agents for cardiovascular diseases.

Journal of nanobiotechnology·2026

Same author

Mechanism of Precipitation Formation and Solubilization Strategy Research in Baihe Gujin Oral Liquid.

Biomedical chromatography : BMC·2026

Same author

Preparation and pharmacological study of <i>Cnidium monnieri</i> volatile oil cream.

Biomedical reports·2026

Same author

Multi-omics reveals the effects and mechanisms of Xinxue Granules against idiopathic pulmonary fibrosis through activating the cAMP/EPAC/CREB axis.

Journal of ethnopharmacology·2026

Same author

Study on the percutaneous absorption and synergistic anti-rheumatoid arthritis effect of the Zanthoxylum bungeanum essential oil on Qin Jiao-Du Huo.

Fitoterapia·2026

Search research articles

相关实验视频

Updated: May 10, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

轻量级多级聚合变压器,用于强大的医疗图像细分.

Xiaoyan Wang¹, Yating Zhu¹, Ying Cui¹

¹School of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, 310023, China; Zhejiang Key Laboratory of Visual Information Intelligent Processing, Hangzhou, 310023, China.

Medical image analysis

|April 25, 2025

概括

MA-TransformerV2提供精确的医疗图像细分,具有轻量级的混合设计. 这种新的方法平衡了有效性和效率,在准确性和模型容量方面超过了现有的方法.

关键词:

卷积卷积是指卷积的过程.可以解释性解释性轻量化轻量化轻量化轻量化轻量化医疗图像细分医疗图像细分多个阶段的聚合聚合.变压器变压器变压器

更多相关视频

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

相关实验视频

Last Updated: May 10, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

科学领域:

医疗成像医学成像
计算机视觉计算机视觉
人工智能的人工智能

背景情况:

准确的医学图像细分需要捕捉多个尺度的特征.
结合卷积神经网络 (CNN) 和变压器的混合网络提供了互补的好处.
现有的混合型号面临着高计算成本或轻量级版本中低于最佳性能的挑战.

研究的目的:

为高效准确的医疗图像细分提出一个强大,轻量级,多阶段的混合架构.
解决当前方法在平衡网络复杂性,计算成本和细分性能方面的局限性.
通过渐进的聚合来增强多尺度特征的提取,以处理复杂的图像变化.

主要方法:

推出了多级聚合变压器版本2 (MA-TransformerV2),是一种新的轻量级多级混合架构.
在每个阶段的双分支编码器中采用平行轻型变压器 (Trans) 块和CNN块.
在瓶中加入了矢量量化块,以分辨特征并减少冗余性.

主要成果:

与最先进的方法相比,MA-TransformerV2显示出更高的细分精度和模型容量.
提出的方法在保持低计算成本的同时实现了高性能.
在公共数据集上的实验结果验证了MA-TransformerV2架构的有效性和效率.

结论:

在MA-TransformerV2有效地提取多尺度的特征使用混合CNN-变压器的方法与渐进的聚合.
该架构在分割精度,计算效率和模型可解释性之间实现了有利的平衡.
这种轻量级的混合型号代表了医疗图像细分技术的重大进步.