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

Consensus Recommendations for Management of Darier Disease: A Practical Approach.

Dermatology and therapy·2026
Same author

Darier disease-A review highlighting new insights from the Darier Disease International Task Force.

Journal of the European Academy of Dermatology and Venereology : JEADV·2026
Same author

Sun protection and skin cancer screening after childhood cancer-A report from the Swiss Childhood Cancer Survivor Study.

Cancer·2026
Same author

European Dermatologists' Perception and Awareness of Teledermatology Use: Results From an International Survey Promoted by the EADV Teledermatology Task Force.

International journal of dermatology·2026
Same author

Reply.

The Journal of allergy and clinical immunology·2026
Same author

Automated triage of cancer-suspicious skin lesions with 3D total-body photography.

NPJ digital medicine·2025

相关实验视频

Updated: Sep 13, 2025

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

2.9K

强大的T-Loss用于医疗图像分割.

Alvaro Gonzalez-Jimenez1, Simone Lionetti2, Philippe Gottfrois1

  • 1Department of Biomedical Engineering, University of Basel, Hegenheimermattweg 167b, Allschwil, 4123, Switzerland.

Medical image analysis
|July 31, 2025
PubMed
概括

医疗图像分割的新损失函数T-Loss有效地处理使用Student-t分布的噪音面罩. 它在皮肤病变和肺部细分任务中表现优于传统方法.

关键词:
深度学习是一种深度学习.标签噪声 标签噪声肺部细分的细分 肺部的细分医疗图像细分 医疗图像细分强大的损失函数功能.皮肤病变细分 皮肤病变细分学生-t 分布 学生-t 分布

更多相关视频

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K

相关实验视频

Last Updated: Sep 13, 2025

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

2.9K
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K

科学领域:

  • 医疗成像医学成像
  • 计算机视觉 计算机视觉
  • 机器学习 机器学习

背景情况:

  • 医学图像细分对于诊断和治疗计划至关重要.
  • 医疗数据集中的噪音标签是一个常见的挑战,影响模型性能.
  • 现有的损失函数与显著的标签噪声作斗争.

研究的目的:

  • 介绍T-Loss,一种用于强大的医疗图像细分的新型损失函数.
  • 解决医疗成像数据集中的噪音口罩的挑战.
  • 提高细分精度和对注释错误的适应性.

主要方法:

  • 从Student-t分布的负日志概率中导出T损失.
  • 使用单个,动态优化的参数来控制对噪音标签的敏感性.
  • 对公共皮肤病变和肺部细分数据集进行了广泛的实验.
  • 模拟各种类型的标签噪声来测试强度.

主要成果:

  • 在皮肤病变和肺部细分任务上的Dice分数中,T-Loss显著超过了传统的损失函数.
  • 对模拟的标签噪声表现出了显著的弹性,模仿人类注释错误.
  • 展示了T-Loss参数在训练期间防止噪音记忆的适应性.

结论:

  • T-Loss提供了一种强大而有效的解决方案,用于用噪音标签对医疗图像进行细分.
  • 它的自适应参数控制使其成为现实世界医学成像应用的有希望的替代方案.
  • 拟议的方法在存在数据缺陷的情况下提高了细分的准确性和可靠性.