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

Stroke: Introduction and Types01:29

Stroke: Introduction and Types

55
A stroke is an acute neurological event caused by the sudden disruption of cerebral blood flow, leading to rapid loss of neuronal function. Neurons depend on continuous oxygen and glucose supply, so even brief interruptions can cause irreversible injury within minutes. Strokes are classified into ischemic and hemorrhagic types.Ischemic StrokeIschemic strokes are most common and occur due to arterial occlusion, depriving brain tissue of oxygen and nutrients. This leads to energy failure, ionic...
55

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

Lipid Profile Testing and Interpretation.

Circulation·2026
Same author

Orchestrator multi-agent clinical decision support system for secondary headache diagnosis in primary care.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

Low-Density Lipoprotein Cholesterol Lowering With Inclisiran Plus Usual Care in Recent Acute Coronary Syndrome: VICTORION-INCEPTION, a Randomized, Controlled, Open-Label Trial.

Journal of the American Heart Association·2026
Same author

Association Between Discrimination in Health Care and ASCVD Among Middle-Aged and Older Adults in the United States.

Circulation. Population health and outcomes·2026
Same author

Targeted use of large language models for EHR-based computable phenotyping.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

Multimodal passive smartphone sensing in older adults: a guide for clinical scientists based upon an ongoing cohort study.

Innovation in aging·2026

相关实验视频

Updated: Apr 30, 2026

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K

大型语言模型的准确性,以在非结构化的电子健康记录数据中识别中风亚型.

Dylan Owens1, Danh Q Nguyen1, Michael Dohopolski2,3

  • 1Department of Medicine, UT Southwestern Medical Center, Dallas, TX. (D.O., D.Q.N., E.D.P., A.M.N.).

Stroke
|July 25, 2025
PubMed
概括

像GPT-4o这样的大型语言模型可以从临床笔记中准确地分类中风类型. 然而,确定特定的缺血性中风亚型仍然是AI在医疗保健中的挑战.

关键词:
疾病的国际分类.人工智能的人工智能是人工智能.电子健康记录是电子健康记录.血流性中风 血流性中风缺血性中风 中风

更多相关视频

Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
06:37

Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke

Published on: July 14, 2023

976
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

684

相关实验视频

Last Updated: Apr 30, 2026

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K
Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
06:37

Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke

Published on: July 14, 2023

976
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

684

科学领域:

  • 医疗信息学 医疗信息学
  • 人工智能在医学中的应用
  • 临床数据分析 临床数据分析

背景情况:

  • 从电子健康记录中准确地分类中风是很困难的,因为结构化数据有限.
  • 临床文档的手动审查通常需要精确的冲击类型.
  • 这项研究调查了大型语言模型对于自动化中风分类的实用性.

研究的目的:

  • 评估GPT-4o在分类中风类型 (缺血性与出血性) 和缺血性中风亚型中的准确性.
  • 在此任务中评估GPT-4o不同提示策略的性能.
  • 将人工智能驱动的分类与专家手工抽象进行比较.

主要方法:

  • 使用GPT-4o的检索增强生成框架被开发用于中风分类.
  • 来自美国心脏协会的数据被用作黄金标准.
  • 三种提示策略 (零射击思维链,专家指导,基于指令) 在两个卫生系统的EHR数据上进行了测试.

主要成果:

  • 在外部验证中,GPT-4o在分类中风类型 (缺血与出血) 中实现了98%的准确性.
  • 对中风类型分类的敏感性和特异性高 (分别为0.98和0.97).
  • 缺血性中风亚型的表现不同,心血管栓塞的高精度 (0.98特异性),但对于密码性中风 (0.40敏感性) 的精度较低.

结论:

  • GPT-4o在区分缺血性和出血性中风类型方面表现出很高的准确性.
  • 该模型在准确分类各种缺血性中风亚型方面存在局限性.
  • 零射击思维链提示被证明是有效的,需要最小的人类投入.