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

Rare SRY-negative 46,XX disorder of sex development with male phenotype and ectopic gonads: a case report.

Frontiers in endocrinology·2026

Same author

Zero-Calibration MI Decoding via Self-Supervised Representation and Ensemble Learning.

IEEE transactions on bio-medical engineering·2026

Same author

MnO<sub>2</sub>-passivated Co<sub>3</sub>O<sub>4</sub> sonozymes for tumor microenvironment re-activated sonodynamic and chemodynamic enhanced immunotherapy.

Biomaterials·2026

Same author

Anti-laser-jamming imaging strategy for cameras based on correlated double sampling technique.

Optics express·2026

Same author

Benefits of forest therapy for adult mental health: a systematic review and meta-analysis based on the Profile of Mood States (POMS).

Frontiers in psychology·2025

Same author

The role of socioemotional skills and phone dependency in predicting patterns of school engagement and burnout among primary school students.

Scientific reports·2025

Search research articles

相关实验视频

Updated: Jul 1, 2025

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

一个高效,快速的医疗图像细分网络.

Diwei Su, Jianxu Luo, Cheng Fei

IEEE journal of biomedical and health informatics

|March 8, 2024

概括

新的浅层层次变压器SHFormer提供了高效的医疗图像细分. 它实现了与复杂模型相比较的准确性,但参数和计算成本大大降低.

科学领域:

医学图像分析医学图像分析
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对于定量分析至关重要.
目前的深度学习模型,如UNet实现高性能,但是计算密集型.
这限制了它们在资源有限的设备上的使用.

研究的目的:

开发一种轻量级但有效的医疗图像细分模型.
为了减少现有细分网络的计算复杂性和参数数量.
为了保持医疗成像应用的高分段精度.

主要方法:

提出了SHFormer,一个浅层次的层次化变压器架构.
引入了一个空间通道连接模块,用于集中注意力.
开发了一个MLP-D模块,用于轻量级的多尺度特征融合.

主要成果:

在ISIC-2018数据集上,SHFormer表现出与最先进的网络可比的性能.
实现了15倍更少的参数,30倍更低的计算复杂性和5倍更高的推理效率.
在具有相似结果的多胞体数据集上验证了概括性.

更多相关视频

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

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: Jul 1, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

结论:

SHFormer为医疗图像细分提供了一个高效的替代方案.
拟议的轻量级模块可实现高性能,降低计算开销.
SHFormer适合在具有有限计算资源的设备上部署.