将临床叙述和结构化表型与大型语言模型和句子转换器相结合
Jihao Cai1, Guozhuang Li1, Yongxin Yang2
1Department of Orthopaedic Surgery, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China; Beijing Key of Big Data Innovation and Application for Skeletal Health Medical Care, Beijing, 100730, China; Key Laboratory of Big Data for Spinal Deformities, Chinese Academy of Medical Sciences, Beijing, 100730, China.
我们开发了LEAP (LLM-Enhanced Automated Phenotyping),这是一个用于从电子健康记录中提取结构化表型的新框架. LEAP显著提高了用于遗传研究和临床应用的表型数据的准确性和可靠性.
科学领域:
- 计算生物学是一种计算生物学.
- 医疗信息学医学信息学
- 基因组学就是基因组学.
背景情况:
- 结构化的表型对于门德尔乱的诊断和遗传研究至关重要.
- 电子健康记录 (EHR) 包含大量的表型数据,但它在很大程度上是无结构的.
- 现有的自动化表型化方法在临床叙述中与语义变化和上下文信息作斗争.
研究的目的:
- 开发一个先进的自动化表型框架,解决当前深度学习模型的局限性.
- 改进从非结构化的临床文本中提取标准化的人类表型本体学 (HPO) 标识符.
- 增强EHR数据对遗传研究和临床决策支持的有用性.
主要方法:
- 提出了LEAP,这是一个两阶段的框架,结合了用于表型提取的大型语言模型 (LLM) 和用于HPO映射的微调句子转换器.
- 该LLM组件可以处理漫长的临床叙述,而无需文本块化.
- 句子转换器模型在超过530万个实例上进行训练,用于准确和确定性的HPO标识符生成.
主要成果:
- 与现有工具相比,LEAP在现实世界EHR测试集上显示出精度 (19.68%-412.68%) 和F1得分 (44.14%-298.77%) 的相对显著改善.
- 在外部基准指标上取得了强的表现,验证了其通用性.
- 该框架确保了有效和确定性的HPO标识符的输出,克服了LLM的局限性.
结论:
- LEAP提供了一种强大而准确的解决方案,用于从非结构化EHR数据中自动化表型化.
- 该框架增强了表型数据的标准化和可用性,用于下游分析,包括基因优先级.
- LEAP代表了利用人工智能在临床信息学和精准医学方面取得的重大进展.
相关概念视频
Improving Translational Accuracy
Improving Translational Accuracy
Language Development
The critical period for language acquisition suggests that the ability to acquire language is at its peak early in life. As people age, this proficiency decreases. Language development begins very...
Language and Cognition
Stereotype Content Model
Higher Mental Functions of the Brain: Language
Language formation and comprehension take place in the dominant hemisphere. The dominant hemisphere is responsible for understanding the meaning of spoken, written, or sign language, as well as the ability to communicate. For most people, the left hemisphere is the dominant one. The right hemisphere, then, gives tone and emotional context to the...

