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

Carbon Stable Isotope Analysis Method of Marine Dissolved Organic Carbon with Laser Absorption Spectroscopy.

Analytical chemistry·2026
Same author

Efficient underwater 3D shape measurement with robust dual-phase complementary unwrapping.

Optics express·2026
Same author

FAT-Net: Frequency-Domain Attention-Guided Topology-Refinement Network for Coronary Artery Segmentation in Invasive Coronary Angiography.

IEEE journal of biomedical and health informatics·2026
Same author

Pan-cancer analysis reveals the oncogenic and immunomodulatory roles of PTGFRN across human cancers.

Scientific reports·2026
Same author

Combining laser-induced breakdown spectroscopy (LIBS) and time-gated Raman spectroscopy for underwater ores identification.

Analytica chimica acta·2026
Same author

Adsorption Characteristics of Heavy Metals onto Functionalized Microplastics.

ACS omega·2026

相关实验视频

Updated: Jul 9, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

559

基于多距离特征不相似性引导的全卷积网络的自动化多片细分.

Nan Mu1,2,3, Jinjia Guo4, Rong Wang1,2,3

  • 1College of Computer Science, Sichuan Normal University, Chengdu 610101, China.

Mathematical biosciences and engineering : MBE
|December 5, 2023
PubMed
概括

这项研究引入了一种新的深度学习方法,用于自动细分结直肠息肉,通过关注特征差异来提高检测准确度,并提高模型性能,以更好地预防癌症.

关键词:
完全卷积网络完全卷积网络.混合损失模块 混合损失模块多距离差异模块的多距离差异模块多层特征减去的多层特征.聚合物细分的聚合物细分

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 9, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

559
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:

  • 医学成像和人工智能 医学成像和人工智能
  • 胃肠道学和瘤学

背景情况:

  • 结肠直肠多,是恶性瘤的前体,通常通过结肠镜检测.
  • 精确的聚细分是具有挑战性的,因为它们的变性特征和周围结构.
  • 现有的卷积神经网络 (CNN) 模型在小息肉和上下文相似性方面扎,导致错过或错误的检测.

研究的目的:

  • 开发一种新,准确的自动多片细分方法.
  • 为了解决当前CNN模型在捕获多特征和细节方面的局限性.

主要方法:

  • 引入一个多距离特征不相似性引导的全卷积网络.
  • 纳入使用多层特征减去 (MLFS) 的多距离差异 (MDD) 模块,以增强特征提取.
  • 利用混合损失 (HL) 模块来监督特征地图并提高预测准确性.

主要成果:

  • 提出的方法在自动聚细分方面表现出卓越的性能.
  • 在四个数据集的六个评估标准中,超过了五种最先进的方法.
  • 有效地捕获不同网络层的区分特征和互补信息.

结论:

  • 这种新型网络有效地细分结直肠多,克服了多变异性和上下文结构带来的挑战.
  • 该MDD和HL模块显著有助于改善特征表示和预测准确性.
  • 这种方法有望提高在结肠镜检查中自动检测的多,帮助早期诊断癌症.