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

相关实验视频

Updated: Jul 4, 2025

An Orthotopic Bladder Tumor Model and the Evaluation of Intravesical saRNA Treatment
08:43

An Orthotopic Bladder Tumor Model and the Evaluation of Intravesical saRNA Treatment

Published on: July 28, 2012

14.8K

一种基于注意力机制的轻量级膀瘤细分方法.

Xiushun Zhao1, Libing Lai2, Yunjiao Li1

  • 1School of Automation, Guangdong University of Technology, Guangzhou, 510006, China.

Medical & biological engineering & computing
|February 2, 2024
PubMed
概括

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

Risk factor analysis for septic shock in female patients after mini-percutaneous nephrolithotripsy.

Translational andrology and urology·2026
Same author

Correction: A novel preoperative risk score for predicting Urosepsis after percutaneous nephrolithotomy: validation and clinical application.

Urolithiasis·2026
Same author

Tooth-Bone Integrated Organoids via Bioactive Glass Mediated Dual Interface Bonding and Rapid Osteogenesis.

Advanced healthcare materials·2026
Same author

A novel preoperative risk score for predicting Urosepsis after percutaneous nephrolithotomy: validation and clinical application.

Urolithiasis·2025
Same author

METS-IR and SII as mediators in the association between smoking and depressive symptoms: insights from NHANES (2005-2018).

BMC psychiatry·2025
Same author

Deubiquitinating enzyme USP42 promotes breast cancer progression by inhibiting JNK/p38-mediated apoptosis.

Scientific reports·2025

一个新的细分网络,NAFF-Net,可以在内镜图像中准确识别膀瘤. 这种方法提高了诊断准确度和治疗规划,通过有效地细分具有模糊边界和可变形状的瘤.

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:

  • 在内镜图像中精确的膀瘤细分对于临床诊断和治疗计划至关重要.
  • 挑战包括模糊的边界和高度可变的瘤形状,使细分工作复杂化.

研究的目的:

  • 在内镜成像中开发先进的细分网络,以改善膀瘤检测.
  • 为了提高诊断准确度,并促进更好的患者治疗策略.

主要方法:

  • 提出了一个嵌套的注意力特征融合细分网络 (NAFF-Net),使用编码器-解码器架构.
  • 整合了一个加权的金字塔聚合模块 (WPPM) 带有状卷积,用于增强特征提取,以及一个嵌套的注意力特征融合 (NAFF) 模块,用于细节聚焦.
  • 开发了一个加权的混合损失函数来解决阶级失衡问题.

主要成果:

  • NAFF-Net实现了优异的细分性能,其平均交叉在欧盟 (MIoU) 的84.05%,MP精度为91.52%,MRecall为90.81%,F1得分为91.16%.
  • 在公开数据集Kvasir-SEG和CVC-ClinicDB上表现出强的结果.
  • 与其他模型相比,参数数量减少,提高了部署能力.

结论:

关键词:
值得注意的是融合特征.膀瘤的细分 膀瘤的细分权重混合损失函数的权重.有权重的金字塔聚合模块.

更多相关视频

A Murine Orthotopic Bladder Tumor Model and Tumor Detection System
06:23

A Murine Orthotopic Bladder Tumor Model and Tumor Detection System

Published on: January 12, 2017

14.8K
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

相关实验视频

Last Updated: Jul 4, 2025

An Orthotopic Bladder Tumor Model and the Evaluation of Intravesical saRNA Treatment
08:43

An Orthotopic Bladder Tumor Model and the Evaluation of Intravesical saRNA Treatment

Published on: July 28, 2012

14.8K
A Murine Orthotopic Bladder Tumor Model and Tumor Detection System
06:23

A Murine Orthotopic Bladder Tumor Model and Tumor Detection System

Published on: January 12, 2017

14.8K
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
  • 在膀瘤细分精度和效率方面,NAFF-Net提供了显著的进步.
  • 该网络处理复杂图像特征的能力及其计算效率使其适合临床部署.