Search research articles

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 Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

Improving Translational Accuracy

Improving Translational Accuracy

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

Automating infection indicator extraction in home healthcare through instruction-tuned large language models.

Journal of the American Medical Informatics Association : JAMIA·2026

Same author

Privacy-preserving verification of preprocessing in federated learning for genomic data.

JAMIA open·2026

Same author

Explainable drug side effect prediction in central neural system via biologically informed graph neural network.

Translational psychiatry·2026

Same author

Privacy-Preserving Verification of ML Preprocessing via Model Behavior Indicators.

IEEE transactions on privacy·2026

Same author

Information extraction from clinical notes: are we ready to switch to large language models?

Journal of the American Medical Informatics Association : JAMIA·2026

Same author

A medically grounded LLM agent-based tool to detect patient safety events in medical records: Identifying patient safety events with large language models.

medRxiv : the preprint server for health sciences·2025

Search research articles

相关实验视频

Updated: Jul 4, 2025

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

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

Published on: December 6, 2024

通过快速工程改进临床命名实体识别的大型语言模型.

Yan Hu¹, Qingyu Chen^2,3, Jingcheng Du¹

¹McWilliams School of Biomedical Informatics, Houston, TX, United States.

Journal of the American Medical Informatics Association : JAMIA

|January 28, 2024

概括

像GPT-4这样的大型语言模型 (LLM) 对临床命名实体识别 (NER) 任务显示出希望. 特定任务提示显著提高了LLM的性能,减少了在医疗保健中需要大量注释数据的需求.

关键词:

在 GPT-3.5 中使用.在 GPT-4 中使用.临床命名实体认可大型语言模型.快速的工程迅速的工程

更多相关视频

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

相关实验视频

Last Updated: Jul 4, 2025

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

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

Published on: December 6, 2024

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

科学领域:

人工智能的人工智能
自然语言处理自然语言处理.
生物医学信息学生物医学信息学

背景情况:

大型语言模型 (LLM) 显示了处理复杂临床数据的潜力.
临床命名实体识别 (NER) 对于从电子健康记录中提取信息至关重要.
当前的方法往往需要大量的注释数据集,限制了可扩展性.

研究的目的:

评估GPT-3.5和GPT-4在临床NER任务上的性能.
开发和评估特定任务的快速框架,以提高临床环境中的LLM能力.
为了比较LLM的表现与已建立的模型,如BioClinicalBERT.

主要方法:

进行了两个临床NER任务:从MTSamples提取概念,并从VAERS识别不良事件.
开发了一个提示框架,包括基线,指导方针,基于错误分析和少量学习提示.
用放松的F1分数评估模型性能,并与BioClinicalBERT进行比较.

主要成果:

通过特定任务的提示框架,GPT-3.5和GPT-4的性能显著提高.
使用所有提示组件,GPT-4获得了0.861 (MTSamples) 和0.736 (VAERS) 的F1分数.
虽然LLM的表现没有超过BioClinicalBERT,但它在最小的培训数据下显示出有希望的结果.

结论:

特定任务提示提高了临床NER的LLMs的可行性.
实际上,LLM,特别是GPT-4,显示出接近最先进的性能与精心及时工程的潜力.
需要进一步的研究和完善的评估方案,才能在临床应用中充分利用LLM.