神经历史既结合并被生成的人工智能查询
Jung-Hyun Lee1,2,3, Eunhee Choi4, Sergio L Angulo1,2
1Department of Neurology, State University of New York Downstate Health Sciences University, Brooklyn, NY, United States.
像GPT-4这样的大型语言模型 (LLM) 显示出改善病史记录的前景. 这项研究发现,检索患者病史细节的准确率为81%,有助于差异诊断.
科学领域:
- 人工智能在医学中的应用
- 临床信息学 临床信息学
- 自然语言处理自然语言处理.
背景情况:
- 大型语言模型 (LLM) 提供了增强医疗记录系统和患者互动的潜力.
- 目前的方法,如等待室问卷,可能是低效的.
- 数字双胞胎和医疗保健对话代理 (HCAs) 是新兴的概念.
研究的目的:
- 评估GPT-4用于初始病史记录的使用.
- 评估基于LLM的系统在提取患者病史数据中的准确性.
- 探索LLMs在产生差异诊断方面的潜力.
主要方法:
- 使用已发表的病例报告 (头痛,中风,神经退行性疾病) 的观察试点研究.
- 采用了三个GPT-4模型:一个患者数字双胞胎 (P),一个神经科医生模型 (N) 查询P,以及一个监督模型 (S) 合成对话.
- 每个案例都被分析了五次,以确保可靠性和一致性.
主要成果:
- 目前疾病史 (HPI) 内容检索的整体准确率为81%.
- 具体准确率包括头痛的84%,中风的82%,神经退行性疾病的77%.
- 在LLM产生的差异诊断排名在第89个百分点.
结论:
- 三方LLM模型在从医疗病例报告中提取关键信息方面表现出显著的准确性.
- 需要通过电子病历 (EMR) HPI进行进一步验证,并与患者直接互动.
- 未来的应用可能涉及诊断数字双胞胎,集成实时健康监测数据.
更多相关视频
08:43Application of Granger Causality Analysis of the Directed Functional Connection in Alzheimer's Disease and Mild Cognitive Impairment
Published on: August 7, 2017
09:47Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
相关概念视频
Reason and Intuition
Neural Circuits
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neurogenesis and Regeneration of Nervous Tissue
Disorders of the Nervous Tissue
Homeostatic Imbalances:
Alzheimer's disease manifests as a gradual decline in memory and cognitive abilities, attributed to the buildup of amyloid plaques and neurofibrillary tangles in the brain.
Parkinson's disease arises from the...
Spinal Cord: Information Processing
Sensory Information Processing
Sensory information processing begins at the sensory receptors located in the skin and other tissues, which detect somatic sensory stimuli such as touch, temperature, or pain. These receptors function as catalysts, initiating...
Traumatic Brain Injury l: Introduction
