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

Actigraphy Meets AI: A Digital Biomarker for Parkinson's Disease and Isolated REM Sleep Behaviour Disorder.

Journal of sleep research·2026
Same author

The EU AI Act: implications and compliance guidance for healthcare facilities.

Frontiers in digital health·2026
Same author

PSA persistence in the PSMA PET era: reassessing a risk factor after radical prostatectomy.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026
Same author

Large Language Model-Based Classification of Case Report Abstracts: A Pilot Study on Interactions Between Radiotherapy and Systemic Therapies.

JCO clinical cancer informatics·2026
Same author

Self-supervised representation learning reveals explainable physiological structure in high-dimensional magnetocardiography.

NPJ digital medicine·2026
Same author

Hypocretin-1/ Orexin-A fragment1-16 as a potential surrogate marker for diagnosing narcolepsy type 1.

Sleep·2026

相关实验视频

Updated: Jan 18, 2026

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

1.0K

实施一个资源简单和低代码的大型语言模型系统,从乳房检查报告中提取信息:试点研究

Fabio Dennstädt1,2, Simon Fauser3, Nikola Cihoric3

  • 1Department of Radiation Oncology, Inselspital, Bern University Hospital and University of Bern, Bern, Switzerland. fabio.dennstaedt@insel.ch.

Journal of imaging informatics in medicine
|September 10, 2025
PubMed
概括

开源的大型语言模型 (LLM) 可以从本地硬件上的乳房镜报告中提取数据. 特定任务的提示显著提高了准确性,使LLM可用于临床使用.

关键词:
人工智能的人工智能是人工智能.数据提取数据提取大型语言模型.乳房学 乳房学 乳房学自然语言处理自然语言处理.

更多相关视频

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

811
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.3K

相关实验视频

Last Updated: Jan 18, 2026

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

1.0K
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

811
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.3K

科学领域:

  • 医疗成像中的人工智能
  • 在放射学中使用自然语言处理.

背景情况:

  • 大型语言模型 (LLM) 显示出从自由文本放射学报告中提取数据的前景.
  • 目前的研究主要通过API使用LLM,限制可访问性和本地部署.

研究的目的:

  • 评估在有限的本地硬件上使用开源LLM用于从乳房检查报告中提取数据的可行性.
  • 评估不同提示策略对提取性能的影响.

主要方法:

  • 定义了79个常见数据元素 (CDE) 用于乳房镜报告.
  • 使用了五个开源的LLM,可在单个GPU上部署,用于数据提取.
  • 我们比较了五种提示方法,包括默认提示,思维链提示和短暂提示.
  • 使用精度,微回忆,微F1和确定性值分析性能.

主要成果:

  • 高互评价协议 (科恩的卡帕0.83) 建立了基本真理.
  • 默认的LLM提示实现了59.2-72.9%的准确性.
  • 针对特定任务的快速调整提高了准确度,达到64.7-85.3%.
  • 确定性值提高了准确度超过90%,但使覆盖率降低到50%以下.

结论:

  • 开源的LLM对于使用有限的计算资源从乳房镜报告中提取信息是有效的.
  • 模型选择和快速工程对于最佳性能至关重要.
  • 基于CDE的框架提高了数据提取的清晰度和结构.