A large language model–based generative natural language processing framework fine‐tuned on clinical notes accurately extracts headache frequency from electronic health records
Headache: The Journal of Head and Face Pain · 2024
DOI: 10.1111/head.14702 ↗Kurz gefasst
Fachartikel in der Zeitschrift Headache: The Journal of Head and Face Pain, 2024. Verfasst von Chia-Chun Chiang und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neurology, Mayo Clinic, Rochester, Minnesota, USA. Hauptthemen laut MEDLINE: Natural Language Processing, Electronic Health Records. Die Arbeit wurde bisher 43-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS, NIH/National Cancer Institute, National Cancer Institute.
Abstract (Originalsprache)
Objective
To develop a natural language processing (NLP) algorithm that can accurately extract headache frequency from free‐text clinical notes.
Background
Headache frequency, defined as the number of days with any headache in a month (or 4 weeks), remains a key parameter in the evaluation of treatment response to migraine preventive medications. However, due to the variations and inconsistencies in documentation by clinicians, significant challenges exist to accurately extract headache frequency from the electronic health record (EHR) by traditional NLP algorithms.
Methods
This was a retrospective cross‐sectional study with patients identified from two tertiary headache referral centers, Mayo Clinic Arizona and Mayo Clinic Rochester. All neurology consultation notes written by 15 specialized clinicians (11 headache specialists and 4 nurse practitioners) between 2012 and 2022 were extracted and 1915 notes were used for model fine‐tuning (90%) and testing (10%). We employed four different NLP frameworks: (1) ClinicalBERT (Bidirectional Encoder Representations from Transformers) regression model, (2) Generative Pre‐Trained Transformer‐2 (GPT‐2) Question Answering (QA) model zero‐shot, (3) GPT‐2 QA model few‐shot training fine‐tuned on clinical notes, and (4) GPT‐2 generative model few‐shot training fine‐tuned on clinical notes to generate the answer by considering the context of included text.
Results
The mean (standard deviation) headache frequency of our training and testing datasets were 13.4 (10.9) and 14.4 (11.2), respectively. The GPT‐2 generative model was the best‐performing model with an accuracy of 0.92 (0.91, 0.93, 95% confidence interval [CI]) and R 2 score of 0.89 (0.87, 0.90, 95% CI), and all GPT‐2–based models outperformed the ClinicalBERT model in terms of exact matching accuracy. Although the ClinicalBERT regression model had the lowest accuracy of 0.27 (0.26, 0.28), it demonstrated a high R 2 score of 0.88 (0.85, 0.89), suggesting the ClinicalBERT model can reasonably predict the headache frequency within a range of ≤ ± 3 days, and the R 2 score was higher than the GPT‐2 QA zero‐shot model or GPT‐2 QA model few‐shot training fine‐tuned model.
Conclusion
We developed a robust information extraction model based on a state‐of‐the‐art large language model, a GPT‐2 generative model that can extract headache frequency from EHR free‐text clinical notes with high accuracy and R 2 score. It overcame several challenges related to different ways clinicians document headache frequency that were not easily achieved by traditional NLP models. We also showed that GPT‐2–based frameworks outperformed ClinicalBERT in terms of accuracy in extracting headache frequency from clinical notes. To facilitate research in the field, we released the GPT‐2 generative model and inference code with open‐source license of community use in GitHub. Additional fine‐tuning of the algorithm might be required when applied to different health‐care systems for various clinical use cases.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Neurology, Mayo Clinic, Rochester, Minnesota, USA.
- Department of Radiology, Mayo Clinic, Phoenix, Arizona, USA.
- Department of Neurology, Mayo Clinic, Phoenix, Arizona, USA.
- Department of Pharmacy, Mayo Clinic, Rochester, Minnesota, USA.
- Department of Cardiology, Mayo Clinic, Rochester, Minnesota, USA.
- Department of Biomedical Informatics, School of Medicine, Emory University, Atlanta, Georgia, USA.
ORCID-Profile: Chia-Chun Chiang, Man Luo, Gina Dumkrieger, Shubham Trivedi, Yi-Chieh Chen, Chieh-Ju Chao, Todd J Schwedt, Abeed Sarker, Imon Banerjee
Förderung
- NCI NIH HHS
- NIH/National Cancer Institute
- National Cancer Institute
Zitieren
APA
Chiang, C. C., Luo, M., Dumkrieger, G., Trivedi, S., Chen, Y. C., Chao, C. J., Schwedt, T. J., Sarker, A., & Banerjee, I. (2024). A large language model–based generative natural language processing framework fine‐tuned on clinical notes accurately extracts headache frequency from electronic health records. Headache: The Journal of Head and Face Pain. https://doi.org/10.1111/head.14702
Vancouver
Chiang CC, Luo M, Dumkrieger G, Trivedi S, Chen YC, Chao CJ, et al. A large language model–based generative natural language processing framework fine‐tuned on clinical notes accurately extracts headache frequency from electronic health records. Headache: The Journal of Head and Face Pain. 2024. doi:10.1111/head.14702
BibTeX
@article{Chiang2024,
author = {Chiang, Chia-Chun and Luo, Man and Dumkrieger, Gina and Trivedi, Shubham and Chen, Yi-Chieh and Chao, Chieh-Ju and Schwedt, Todd J and Sarker, Abeed and Banerjee, Imon},
title = {A large language model–based generative natural language processing framework fine‐tuned on clinical notes accurately extracts headache frequency from electronic health records},
journal = {Headache: The Journal of Head and Face Pain},
year = {2024},
doi = {10.1111/head.14702}
}
EUVIMED