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

The Interference of <i>Mnsod3</i> Enhances the Tolerance of <i>Pleurotus ostreatus</i> Mycelia to Abiotic Stress by Reshaping the Cell Wall.

Journal of fungi (Basel, Switzerland)·2026
Same author

Efficacy and Safety of a Generic Clevidipine in Hypertensive Urgencies and Non-Fulminant Hypertensive Emergencies: A Phase III, Multicenter, Randomized, Double-Blind, Positive Drug Parallel-Controlled Study.

Journal of clinical hypertension (Greenwich, Conn.)·2025
Same author

Local-global multi-scale attention network for medical image segmentation.

PeerJ. Computer science·2025
Same author

Metabolic Remodeling of the Tricarboxylic Acid Cycle and Glycolysis Reveals Cold-Induced Respiratory Adaptations in <i>Streltzoviella insularis</i> (Staudinger) (Lepidoptera: Cossidae) Larvae.

Insects·2025
Same author

Recombinant monoclonal antibody siltartoxatug versus plasma-derived human tetanus immunoglobulin for tetanus: a randomized, double-blind, active-controlled, phase 3 trial.

Nature medicine·2025
Same author

IRAK4 Targeting: A Breakthrough Approach to Combat Hidradenitis Suppurativa.

Biologics : targets & therapy·2025

相关实验视频

Updated: Jul 8, 2025

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K

基于U-Net的皮肤病变细分方法的研究.

Dapeng Cheng1,2, Jiale Gai1, Yanyan Mao1

  • 1School of Computer Science and Technology, Shandong Technology and Business University, Yantai, 264005, Shandong, China.

Heliyon
|December 11, 2023
PubMed
概括

本研究介绍了EA-Net,这是一种新的卷积神经网络 (CNN),通过结合注意力机制来增强皮肤病变细分. EA-Net提高了准确性,特别是在具有模糊边界的具有挑战性的病例中,有助于临床决策.

关键词:
注意力机制注意力机制卷积神经网络是一种卷积神经网络.功能提取 功能提取图像细分 图像细分 图像细分皮肤病变 皮肤病变

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

2.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

406

相关实验视频

Last Updated: Jul 8, 2025

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K
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

2.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

406

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:

  • 准确的皮肤病变细分对于临床诊断至关重要,但由于位置,形状和规模的变化而受到阻碍.
  • 现有的卷积神经网络 (CNN) 难以保留本地图像特征并突出显示相关信息,限制了它们的临床应用.

研究的目的:

  • 开发一个增强的CNN模型 (EA-Net) 以实现更准确的皮肤病变细分.
  • 改进区域特征的保护和特征地图在细分过程中的相关性.

主要方法:

  • 提出了EA-Net,一个基于U-Net的CNN,在编码器中包含一个像素级的注意模块 (PA),以保存本地特征.
  • 集成了一个空间多尺度注意模块 (SA) 后解码器来完善特征地图并增强空间相关性.
  • 在ISIC 2017和ISIC 2018皮肤病变数据集上评估了EA-Net.

主要成果:

  • 与U-Net相比,EA-Net在两个数据集上都表现出更好的表现.
  • 实现了1.94% (ISIC 2017) 和5.38% (ISIC 2018) 的平均子得分改善.
  • 显示的交叉点在联盟 (IoU) 上增加了2.69%和8.31%,平均对称表面距离 (ASSD) 减少了0.3783和0.5432像素.

结论:

  • 拟议的EA-Net通过整合注意力机制,有效地提高了皮肤病变细分的准确性.
  • 该模型在细分具有模糊边界和复杂条件的病变方面表现出色,证明了在医学图像分析中关注的价值.