Leveraging large language models to extract smoking history from clinical notes for lung cancer surveillance

Accurate smoking documentation in electronic health records (EHRs) is crucial for risk assessment and patient monitoring. However, key information is often missing or inaccurately recorded. Large language models (LLMs) present a promising solution for interpreting clinical narratives to extract comprehensive smoking data. We developed a framework utilizing LLMs combined with rule-based longitudinal smoothing techniques to enhance data quality. We compared generative LLMs (Gemini-1.5-Flash, PaLM-2-Text-Bison, GPT-4) against BERT-based models using 1683 manually annotated clinical notes from 518 patients across Stanford and Sutter Health systems. Generative LLMs achieved superior performance ( > 96% accuracy) across seven smoking variables, with external validation showing robust generalizability (97.5-98.8% accuracy). We deployed Gemini-1.5-Flash to 79,408 notes from 4792 lung cancer patients, demonstrating that risk model-based surveillance incorporating smoking factors outperformed NCCN Guidelines in identifying second malignancies. Our study highlights the potential of generative LLMs to improve smoking history documentation quality, enhancing lung cancer surveillance and broader clinical applications.

Authors

Institutions

Publication Details

Journal
npj Digital Medicine
Published
2025-11-28
DOI
https://doi.org/10.1038/s41746-025-02009-y
Citations
4
Primary Topic
Machine Learning in Healthcare
Type
article
Field-Weighted Citation Impact
9.92

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Leveraging large language models to extract smoking history from clinical notes for lung cancer surveillance

Tony Chen, Fatma Güntürkün, Rajnish Kaushik, Tae Yoon Lee et al.
4 citations
npj Digital Medicine
Machine Learning in Healthcare
9.92
article

Leveraging large language models to extract smoking history from clinical notes for lung cancer surveillance

Tony Chen, Fatma Güntürkün, Rajnish Kaushik, Tae Yoon Lee, Eunji Choi, Manisha Desai, Joel W. Neal, Heather A. Wakelee, Chloe C. Su, Ann N. Leung, Su‐Ying Liang, A. Solomon Henry, Mengrui Zhang, Anna Graber‐Naidich, Archana Bhat, Bohong Gu, Allison W. Kurian, Aparajita Khan, Timothy John Ellis-Caleo, Summer S. Han, Leah M. Backhus, Natalie S. Lui, Curtis P. Langlotz, Ingrid Luo, Victoria Y. Ding, Mina Satoyoshi, Grant M. Nieda
article en
4 citations

Abstract

Accurate smoking documentation in electronic health records (EHRs) is crucial for risk assessment and patient monitoring. However, key information is often missing or inaccurately recorded. Large language models (LLMs) present a promising solution for interpreting clinical narratives to extract comprehensive smoking data. We developed a framework utilizing LLMs combined with rule-based longitudinal smoothing techniques to enhance data quality. We compared generative LLMs (Gemini-1.5-Flash, PaLM-2-Text-Bison, GPT-4) against BERT-based models using 1683 manually annotated clinical notes from 518 patients across Stanford and Sutter Health systems. Generative LLMs achieved superior performance ( > 96% accuracy) across seven smoking variables, with external validation showing robust generalizability (97.5-98.8% accuracy). We deployed Gemini-1.5-Flash to 79,408 notes from 4792 lung cancer patients, demonstrating that risk model-based surveillance incorporating smoking factors outperformed NCCN Guidelines in identifying second malignancies. Our study highlights the potential of generative LLMs to improve smoking history documentation quality, enhancing lung cancer surveillance and broader clinical applications.

npj Digital MedicineVol. 8(1)
Sutter Health (US), Cornell University (US), Stanford Health Care (US), Stanford Medicine (US), Weill Cornell Medicine (US), Indian Institute of Technology BHU (IN), Banaras Hindu University (IN), Stanford University (US)
National Institutes of Health, National Cancer Institute
Openalex Percentile: Top 2%
Machine Learning in Healthcare
9.92
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.