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

Towards a Reliable and Rapid Automated Grading System in Facial Palsy Patients: Facial Palsy Surgery Meets Computer Science.

Journal of clinical medicine·2022
Same author

Correction to: New aspects in digital breast assessment: further refinement of a method for automated digital anthropometry.

Archives of gynecology and obstetrics·2021
Same author

New aspects in digital breast assessment: further refinement of a method for automated digital anthropometry.

Archives of gynecology and obstetrics·2020
Same journal

Contactless robotic system for linear catheter advancement using magnetic actuation.

International journal of computer assisted radiology and surgery·2026
Same journal

Sound source localization for spatial mapping of surgical actions in dynamic scenes.

International journal of computer assisted radiology and surgery·2026
Same journal

ESD-VesNet: uncertainty-aware vessel segmentation network for endoscopic submucosal dissection with hard negative mining.

International journal of computer assisted radiology and surgery·2026
Same journal

Lean Unet: a compact model for image segmentation.

International journal of computer assisted radiology and surgery·2026
Same journal

Strain alignment: toward assessing mechanical plausibility of predicted displacement fields.

International journal of computer assisted radiology and surgery·2026
Same journal

Vascular geometry characterization for AI-based endovascular navigation.

International journal of computer assisted radiology and surgery·2026
查看所有相关文章

相关实验视频

Updated: Jun 12, 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

2.7K

提高零射击多标签内镜仪器分类中的概括性.

Raphaela Maerkl1, Tobias Rueckert2,3, David Rauber2

  • 1Regensburg Medical Image Computing (ReMIC), OTH Regensburg, 93053, Regensburg, Germany. raphaela.maerkl@st.oth-regensburg.de.

International journal of computer assisted radiology and surgery
|June 11, 2025
PubMed
概括
此摘要是机器生成的。

改善医疗AI的零射击学习,本研究使用句子嵌入和z-score正常化来增强未见的手术仪器的识别,显著提高准确性.

关键词:
一般化的零射击学习.多个标签分类的分类.句子嵌入的句子嵌入.手术仪器 手术仪器是指手术仪器.Z-分数规范化的标准化

更多相关视频

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

2.4K
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

483

相关实验视频

Last Updated: Jun 12, 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

2.7K
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

2.4K
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

483

科学领域:

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

背景情况:

  • 神经网络难以将其推广到看不见的类,这是安全关键医疗应用中的一个关键问题.
  • 零射击学习 (ZSL) 通过利用语义数据提供解决方案,但性能取决于嵌入质量.

研究的目的:

  • 调查ZSL在医疗图像识别中的完整描述性句子嵌入与简单单词嵌入的有效性.
  • 评估z-score规范化对未见类嵌入性能的影响.

主要方法:

  • 使用 Sentence-BERT 来生成描述性句子嵌入作为类表示.
  • 将句子嵌入与BERT衍生的词嵌入进行比较.
  • 应用z-score规范化作为后处理步骤.
  • 在多标签的通用零射击学习任务上进行评估,用于内镜图像中的手术仪器识别.

主要成果:

  • 将句子嵌入与z分数正常化相结合,显著提高了未见类的性能.
  • 接收器操作特征曲线 (AUROC) 下面的面积,未见的类从43.9%增加到64.9%.
  • 不见类的多标签精度从26.1%上升到79.5%.

结论:

  • 句子嵌入和z-score规范化大大提高了零射击学习模型的概括能力.
  • 拟议的方法有望提高医疗AI应用中的可靠性.
  • 建议在不同的数据集和领域进行进一步验证,以确认可靠性.