使用人工智能的健康加速器计划的社会决定因素的内容分析:公共卫生从业人员的用例
Kelli DePriest1, John Feher Iii, Kailen Gore
1Author Affiliations: Research Triangle Institute: RTI International, Research Triangle Park, North Carolina (Dr DePriest, Mr Feher III, Ms Gore, Dr Glasgow, and Mr Chew); Association of State and Territorial Health Officials (ASTHO), Arlington, Virginia (Mr Grant); National Association of County and City Health Officials (NACCHO), Washington, District of Columbia (Mr Holtgrave); Centers for Disease Control and Prevention (CDC), Atlanta, Georgia (Dr Hacker).
大型语言模型 (LLM) 显示出分析公共卫生文件的潜力,但需要谨慎使用. 虽然更快,但人工智能在数据抽象和合成中的准确性低于人类分析,有可能出现错误.
科学领域:
- 公共卫生信息学 公共卫生信息学
- 医疗保健中的人工智能
- 定性数据分析 数据分析.
背景情况:
- 公共卫生实践依赖于分析许多文件,如战略计划和计划评估的需求评估.
- 对这些定性数据源的手动内容分析是耗时的,限制了实时程序改进.
- 大型语言模型 (LLM) 为自动化和加快这种分析提供了潜在的解决方案.
研究的目的:
- 评估使用LLM来分析公共卫生文件的可行性.
- 将GPT-4o的内容分析与人类分析的准确性和效率进行比较.
- 评估从社区健康改善计划中抽象和综合信息.
主要方法:
- 一个研究小组对4个社区健康改善计划进行了内容分析.
- 将GPT-4o的性能与研究小组对20个数据元素的结果进行了比较.
- 评估了资源需求,包括快速工程和错误纠正的时间.
主要成果:
- 在数据抽象方面,GPT-4o实现了79%的准确性,而研究小组的数据抽象率为94%,有伪造数据的情况.
- 对于合成,GPT-4o准确度为50%,在18个元素中有9个错误.
- 虽然GPT-4o最初更快,但当考虑到快速的工程和错误纠正时,整体时间节省会减少.
结论:
- 法律法规可以加快公共卫生文件的分析,但必须仔细考虑准确性.
- 公共卫生专业人员应该权衡潜在的资源节约与数据不准确的风险.
- 建议采取平衡的方法,将人工智能工具与人类监督相结合,以获得最佳的结果.
更多相关视频
相关概念视频
Issues And Trends In Healthcare Delivery System
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
Methods Of Healthcare Delivery System
Managed Care System:
The managed care system is designed to control the cost while maintaining the quality of care. The patient's care from admission to discharge is planned by the primary care provider or the case manager, also known as the gatekeeper. In a managed care system, the number of care providers is...
Primary Healthcare Services
In 1978, international leaders convened in Alma-Ata, Kazakhstan, for what would be a pivotal event in global health. The Alma-Ata Declaration was the first to call...
Preventive Healthcare Services
Models of Health Promotion and Illness Prevention I
The health belief model (HBM) attempts to predict health-related behavior in specific belief patterns. According to the HBM, a person's...
Steps in Outbreak Investigation


