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

Endoscopic Procedures III: Video Capsule Endoscopy01:28

Endoscopic Procedures III: Video Capsule Endoscopy

653
Capsule endoscopy, or wireless or video capsule endoscopy, is a diagnostic procedure for examining the entire gastrointestinal tract. Patients swallow a capsule about the size of a vitamin tablet. The capsule is equipped with a transmitter, a battery, an LED light source, and a color video camera to capture images throughout the gastrointestinal tract. This procedure is particularly useful for diagnosing conditions such as Crohn's disease, ulcerative colitis, tumors, polyps, ulcers,...
653

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

Clinical evaluation of a computer-assisted decision support and documentation system for the primary care of polytrauma patients.

Frontiers in digital health·2026
Same author

Feasibility of Vibroacoustic Sensing for Detection of Peritoneal Entry During Laparoscopic Access: A Pilot Study in a Human Body Donor.

Diagnostics (Basel, Switzerland)·2026
Same author

Publisher Correction: Unsupervised risk factor identification across cancer types and data modalities via explainable artificial intelligence.

NPJ digital medicine·2026
Same author

Impact of Renal Impairment and Lymphodepletion Regimen on Outcomes after CAR T Cell Therapy in Relapsed/Refractory Multiple Myeloma

Transplantation and cellular therapy·2026
Same author

Unsupervised risk factor identification across cancer types and data modalities via explainable artificial intelligence.

NPJ digital medicine·2026
Same author

Body composition predicts poor outcomes and reveals immunometabolic dysfunction via single-cell profiling in anti-BCMA CAR T-treated myeloma.

HemaSphere·2026

相关实验视频

Updated: Jan 7, 2026

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

3.3K

使用深度视觉语言模型可以提高内镜耳鼻喉外科手术的辅助应用中的多任务性能.

Richard Bieck1, Martin Sorge2, Katharina Heuermann2

  • 1Innovation Center Computer Assisted Surgery (ICCAS), Leipzig University, Semmelweisstraße 14, 04103, Leipzig, Germany. Richard.bieck@medizin.uni-leipzig.de.

International journal of computer assisted radiology and surgery
|December 22, 2025
PubMed
概括

本研究介绍了用于内镜耳鼻喉手术的视觉语言模型 (VLM),增强图像分类和报告生成. VLM集成视觉和文本数据,优于现有的多任务辅助模型.

关键词:
深度学习是一种深度学习.可以解释的可解释性.这就是FESSESS.图像嵌入式 图像嵌入式基于图像的内镜导航预培训 预培训 预培训嵌入文本 嵌入文本变压器 变压器 变压器视觉语言模型 视觉语言模型

更多相关视频

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
09:41

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery

Published on: May 20, 2016

12.7K
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

980

相关实验视频

Last Updated: Jan 7, 2026

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

3.3K
A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
09:41

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery

Published on: May 20, 2016

12.7K
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

980

科学领域:

  • 医疗人工智能 医疗人工智能
  • 计算机视觉 计算机视觉
  • 自然语言处理自然语言处理.

背景情况:

  • 目前用于内镜辅助的深度学习模型主要使用基于图像的任务.
  • 自然语言处理的整合是有限的,阻碍了全面的援助能力.

研究的目的:

  • 开发和评估一种视觉语言模型 (VLM),用于内镜耳鼻喉手术中的多任务学习.
  • VLM的目标是执行图像分类,文本预测和手术报告生成.

主要方法:

  • 采用了一个VLM架构,用于图像和文本嵌入的域偏向编码器.
  • 该模型在30个内镜手术 (13万张图像,报告) 的新多任务数据集上进行了训练.
  • 根据基线,EndoVit和SurgicalGPT模型对两个VLM变异进行了评估,使用精度,回忆,F1得分,BLEU-2,ROUGE-L和等号相似性.

主要成果:

  • 该VLM提高了图像分类F1得分高达12%,文本生成高达14%.
  • 域特定的VLM略高于EndoVit和外科GPT.
  • 除研究表明,视觉组件有利于语言任务,而文本对地标检测的影响最小.

结论:

  • 开发了一种新的VLM用于内镜耳鼻喉辅助,集成图像和文本数据.
  • VLM取代了三个孤立的模型,提供多任务辅助,并超越了以前的通用基线.
  • 未来的工作需要解决不平衡的阶级分布,并改进结构化文本生成.