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

Secondary hemophagocytic lymphohistiocytosis in pediatric patients: a single-center experience.

Frontiers in pediatrics·2026
Same author

Utilizing [<sup>18</sup>F]-FDG PET/CT Imaging for Enhanced Staging and Treatment Decisions in Pediatric Rhabdomyosarcoma.

Cancers·2026
Same author

Beyond the 5-Year Window: Late-Onset Ocular Morbidity and a Proposed 10-Year Functional Survivorship Protocol for Pediatric Orbital Rhabdomyosarcoma.

Cancers·2026
Same author

Ruptured Wilms Tumor: Clinical Features, Diagnostic Challenges, and Survival Outcomes.

Current oncology (Toronto, Ont.)·2026
Same author

Exploring Two Decades of Cancer Trends in Adolescents and Young Adults: Insights From a Resource-Restricted Country.

World journal of oncology·2026
Same author

Wilms Tumor with Inferior Vena Cava Thrombus: Comparative Analysis of Clinical Characteristics and Outcomes.

Current oncology (Toronto, Ont.)·2026

相关实验视频

Updated: May 29, 2025

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.0K

评估GPT模型用于临床笔记非识别.

Bayan Altalla'1,2, Sameera Abdalla3, Ahmad Altamimi3

  • 1King Hussein Cancer Center, Queen Rania Street, Amman, Jordan. bayan.ahmad995b@gmail.com.

Scientific reports
|January 31, 2025
PubMed
概括
此摘要是机器生成的。

与GPT-3.5.5相比,GPT-4在消除临床注释的识别和生成合成健康数据方面表现出色,显著改善了患者隐私和研究数据的实用性.

更多相关视频

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.2K
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.4K

相关实验视频

Last Updated: May 29, 2025

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.0K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.2K
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.4K

科学领域:

  • 医疗信息学 医疗信息学
  • 医疗保健中的人工智能
  • 数据 隐私 数据 隐私 数据

背景情况:

  • 医疗保健的数字化需要安全的临床数据管理.
  • 患者隐私是处理敏感健康信息的关键问题.
  • 现有的数据去识别和合成数据生成方法需要优化.

研究的目的:

  • 为了评估GPT-3.5和GPT-4的非识别临床笔记.
  • 评估这些模型在生成合成临床数据方面的性能.
  • 使用API访问和零射击提示工程来优化计算效率.

主要方法:

  • 通过API访问使用了GPT-3.5和GPT-4.
  • 采用零射击快速工程来消除识别和合成数据生成.
  • 使用精度,回忆,F1得分和准确度指标评估模型性能.

主要成果:

  • 在非识别的临床注释中,GPT-4在GPT-3.5上表现优越.
  • GPT-4的精度为0.9925,回忆率为0.8318,F1得分为0.8973,准确度为0.9911.
  • 该研究证实了大型语言模型在医疗数据隐私方面的潜力.

结论:

  • GPT-4是提高临床数据患者隐私的高性能工具.
  • 这项研究为平衡医疗保健中的数据实用性和隐私建立了基准.
  • 这些发现支持使用先进的人工智能来安全管理临床数据和促进研究.