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

Computed Tomography01:10

Computed Tomography

4.4K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
4.4K

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

The Risks and Benefits of Synthetic Medical Images.

Radiology·2026
Same author

Small language models in medicine.

Nature biomedical engineering·2026
Same author

Neuro-symbolic artificial intelligence in medicine.

Nature biomedical engineering·2026
Same author

Evaluation of Large Language Models for Structured Data Extraction From Interstitial Lung Disease Clinical Notes: Comparative Study.

Journal of medical Internet research·2026
Same author

Towards Conversational AI for Disease Management.

Nature·2026
Same author

Neural Shape Modeling Reveals Early and Progressive Femoral Bone Shape and Cartilage Thickness Changes After Anterior Cruciate Ligament Reconstruction.

medRxiv : the preprint server for health sciences·2026

相关实验视频

Updated: Jun 21, 2025

Computer-Aided Three-Dimensional Visualization in the Treatment of Locally Advanced Thyroid Cancer
03:55

Computer-Aided Three-Dimensional Visualization in the Treatment of Locally Advanced Thyroid Cancer

Published on: June 9, 2023

514

默林:一个视觉语言基础模型用于3D计算机断层扫描.

Louis Blankemeier1,2,3, Joseph Paul Cohen2, Ashwin Kumar2,3

  • 1Department of Electrical Engineering, Stanford University.

Research square
|July 9, 2024
PubMed
概括

默林是一种新的3D视觉语言模型 (VLM),使用EHR数据和报告来解释腹部CT扫描. 这种人工智能工具增强了医疗图像分析和疾病预测,在单个GPU上高效训练.

更多相关视频

Construction of a Realistic, Whole-Body, Three-Dimensional Equine Skeletal Model using Computed Tomography Data
11:09

Construction of a Realistic, Whole-Body, Three-Dimensional Equine Skeletal Model using Computed Tomography Data

Published on: February 25, 2021

3.1K
Scaled Anatomical Model Creation of Biomedical Tomographic Imaging Data and Associated Labels for Subsequent Sub-surface Laser Engraving SSLE of Glass Crystals
07:57

Scaled Anatomical Model Creation of Biomedical Tomographic Imaging Data and Associated Labels for Subsequent Sub-surface Laser Engraving SSLE of Glass Crystals

Published on: April 25, 2017

8.4K

相关实验视频

Last Updated: Jun 21, 2025

Computer-Aided Three-Dimensional Visualization in the Treatment of Locally Advanced Thyroid Cancer
03:55

Computer-Aided Three-Dimensional Visualization in the Treatment of Locally Advanced Thyroid Cancer

Published on: June 9, 2023

514
Construction of a Realistic, Whole-Body, Three-Dimensional Equine Skeletal Model using Computed Tomography Data
11:09

Construction of a Realistic, Whole-Body, Three-Dimensional Equine Skeletal Model using Computed Tomography Data

Published on: February 25, 2021

3.1K
Scaled Anatomical Model Creation of Biomedical Tomographic Imaging Data and Associated Labels for Subsequent Sub-surface Laser Engraving SSLE of Glass Crystals
07:57

Scaled Anatomical Model Creation of Biomedical Tomographic Imaging Data and Associated Labels for Subsequent Sub-surface Laser Engraving SSLE of Glass Crystals

Published on: April 25, 2017

8.4K

科学领域:

  • 医疗成像中的人工智能
  • 放射学和医学信息学
  • 计算机视觉和自然语言处理

背景情况:

  • 在美国,每年有数以百万计的腹部CT扫描需要放射科医生的解释.
  • 目前用于医学成像的AI模型通常仅限于2D数据和简短的报告.
  • 需要先进的人工智能来协助放射科医生,并从复杂的扫描中提取新的见解.

研究的目的:

  • 介绍Merlin,一个用于腹部CT解释的3D视觉语言模型 (VLM).
  • 在没有手动注释的情况下,利用电子健康记录 (EHR) 和放射学报告进行预训练.
  • 增强自动化医疗图像分析,并从CT扫描中提取生理洞察力.

主要方法:

  • 在一个庞大的临床数据集上训练默林:600多万张CT图像,180多万张EHR诊断码和600多万张放射学报告令牌.
  • 在6种任务类型上评估Merlin,包括零射击分类,交叉模式检索,慢性疾病预测,报告生成和3D细分.
  • 对各种CT数据集进行内部和外部验证,包括公共基准.

主要成果:

  • 默林在多个评估任务中与现有的特定任务基线相比表现良好.
  • 该模型在零射击发现和表型分类以及跨模式检索方面取得了强的结果.
  • 在单个GPU上实现了高效的训练,表明了民主化AI模型开发的潜力.

结论:

  • 默林代表了3DVLMs对于腹部CT解释的重大进步.
  • 该模型能够整合EHR数据和非结构化报告,为临床决策支持提供了强大的工具.
  • 计算效率高的培训方法有助于在医疗保健环境中更广泛地采用先进的人工智能.