SmokeBERT:一种基于变压器模型的双向编码器表示,用于从临床叙述中提取定量吸烟史,以改善肺癌查
Yiming Xue1, Yunzheng Zhu2, Luoting Zhuang2
1Department of Statistics and Data Science, University of California, Los Angeles, CA.
JCO clinical cancer informatics
|December 1, 2025
概括
我们开发了SmokeBERT,这是一种专门的NLP工具,可以准确地从临床笔记中提取详细的吸烟史. 这有助于改善疾病风险评估和肺癌查资格识别.
科学领域:
- 自然语言处理 (NLP) 是一种自然语言处理.
- 临床信息学 临床信息学
- 生物医学数据提取
背景情况:
- 吸烟是癌症和其他疾病的主要风险因素.
- 准确的定量吸烟数据 (包年,戒烟时间) 对于风险评估和肺癌查至关重要.
- 现有的NLP工具在从临床叙述中提取颗粒状的吸烟细节方面面临挑战.
研究的目的:
- 开发和评估一种新的NLP模型,SmokeBERT,用于精确地从临床文本中提取定量吸烟史.
- 根据吸烟史,提高识别符合肺癌查条件的患者的准确性.
主要方法:
- 基于变压器 (BERT) 的四个双向编码器表示模型 (BERT,BioBERT,ClinicalBERT,MedBERT) 在临床叙述上进行了微调.
- 模型接受了3,261个提到吸烟史的句子的训练,以提取六个定量变量.
- 选择了ClinicalBERT,对SmokeBERT进行了微调,并对持久和外部数据集进行了验证.
主要成果:
- 与最先进的NLP模型相比,SmokeBERT在吸烟数据提取方面取得了更高的性能 (F1得分:持久的0.97,外部的0.86).
- SmokeBERT准确地确定了符合肺癌查条件的患者 (例如,≥20包年:97%,≤15年戒烟:100%).
- 性能超过了现有的基于规则和大型变压器模型.
结论:
- 微调的BERT模型SmokeBERT有效地从临床数据中提取详细的定量吸烟史.
- 该模型显示了提高临床风险评估和患者查的巨大潜力.
- 未来的研究将专注于更大的数据集和多语言,语言不可知版本.
相关概念视频
01:25Statistical Methods for Analyzing Epidemiological Data
875
Epidemiological data primarily involves information on specific populations' occurrence, distribution, and determinants of health and diseases. This data is crucial for understanding disease patterns and impacts, aiding public health decision-making and disease prevention strategies. The analysis of epidemiological data employs various statistical methods to interpret health-related data effectively. Here are some commonly used methods:
875
01:11Observational Studies
10.7K
Observational studies are a type of analytical study where researchers observe events without any interventions. In other words, the researcher does not influence the response variable or the experiment's outcome.
There are three types of observational studies – Prospective, retrospective, and cross-sectional.
Prospective Study
Prospective studies, also known as longitudinal or cohort studies, are carried out by collecting future data from groups sharing similar characteristics. One...
There are three types of observational studies – Prospective, retrospective, and cross-sectional.
Prospective Study
Prospective studies, also known as longitudinal or cohort studies, are carried out by collecting future data from groups sharing similar characteristics. One...
10.7K
01:22Chronic Obstructive Pulmonary Disease
2.2K
COPD is defined as a heterogeneous lung condition marked by persistent respiratory symptoms such as dyspnea, cough, and sputum production, caused by abnormalities in the airways that cause airflow obstruction.
Smoking is a primary risk factor for COPD, with over 80% of patients having a history of it. Patients typically experience progressive dyspnea or labored breathing, frequent coughing, and recurrent pulmonary infections. Many eventually succumb to respiratory failure, characterized by...
Smoking is a primary risk factor for COPD, with over 80% of patients having a history of it. Patients typically experience progressive dyspnea or labored breathing, frequent coughing, and recurrent pulmonary infections. Many eventually succumb to respiratory failure, characterized by...
2.2K