Preprint · 2026 · Open access
Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data
arXiv (Cornell University) · Published 6 March 2026 · DOI 10.48550/arxiv.2603.23515
In brief
“Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data” is a 2026 preprint by John Cook and 13 co-authors, published in arXiv (Cornell University). OpenAlex counts 1 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).
- Citations
- 1
- Most cited in
- 2026 (1)
- References
- 14
- Authors
- 14
Abstract
Improving the accuracy and reliability of medical coding reduces clinician burnout and supports revenue cycle processes, freeing providers to focus more on patient care. However, automating the assignment of ICD-10-CM and CPT codes from clinical documentation remains a challenge due to heterogeneous records, nuanced coding guidelines, and long-tail distributions. Large language models have been proposed to help or automate specific medical coding tasks. However, foundation models are not explicitly trained for medical coding and zero-shot coding has yielded poor results. We investigate whether a modern open-weight foundation model can be adapted for an expert-level medical coding task using privacy-preserving synthetic training data derived from electronic health records. We fine-tune Llama 3-70B on pairs of clinical notes and gold codes generated from EHR-grounded templates and coding policies, then evaluate exact-code prediction for ICD-10-CM and CPT. A zero-shot baseline with the unadapted model achieved an F1 score of 0.18 for exact code match. After fine-tuning on the synthetic corpus, exact-match F1 exceeded 0.70, representing a large absolute gain across both code systems. Notably, performance remained high on complex categories that often require multi-step clinical reasoning and code composition, including Advanced Illness and Frailty classes, and the model retained its performance on medical comprehension tasks. These results indicate that synthetic, policy-aware data can efficiently teach a general-purpose large language model to support precise medical coding without exposing protected health information. The approach offers a practical path for training coding agents safely and iteratively on specific tasks that represent real-world populations.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Cook, J., Wyatt, M., Wei, P., Chin, I., Gupta, S., van Vuuren, V. Z., Siburian, R., Spicer, A., Viviano, K., Cami, A., Malhotra, R., Yao, Z., Rasley, J., & Kaushik, G. (2026). Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2603.23515
MLA 9
Cook, John, et al. “Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2603.23515.
Chicago
Cook, John, Michael Wyatt, Peng Wei, Iris Chin, Santosh Gupta, Van Zyl van Vuuren, Richie Siburian, et al. 2026. “Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2603.23515.
Harvard
Cook, J. et al. (2026) ‘Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data’, arXiv (Cornell University). doi:10.48550/arxiv.2603.23515.
IEEE
J. Cook et al., “Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2603.23515.
BibTeX
@misc{cook2026training,
title = {{Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data}},
author = {Cook, John and Wyatt, Michael and Wei, Peng and Chin, Iris and Gupta, Santosh and van Vuuren, Van Zyl and Siburian, Richie and Spicer, Amanda and Viviano, Kristen and Cami, Alda and Malhotra, Raunaq and Yao, Zhewei and Rasley, Jeff and Kaushik, Gaurav},
howpublished = {arXiv (Cornell University)},
year = {2026},
doi = {10.48550/arxiv.2603.23515},
url = {https://doi.org/10.48550/arxiv.2603.23515}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Health Sciences › Health Professions › Health Information Management › Medical Coding and Health Information
- medical coding
- ICD-10-CM
- CPT codes
- large language models
- Llama 3
- fine-tuning
- synthetic health data
- privacy-preserving data generation
Work with this paper
Most-cited papers in Health Professions
- DETERMINATION OF SERUM PROTEINS BY MEANS OF THE BIURET REACTIONAllan Godfrey Gornall, Charles J. Bardawill, Maxima M. David · Journal of Biological Chemistry · 1949 · 16,172 citations
- Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation scienceLaura J. Damschroder, David Clark Aron, Rosalind E. Keith and 3 more · Implementation Science · 2009 · 15,527 citations
- FUNCTIONAL EVALUATION: THE BARTHEL INDEX.FLORENCE I. MAHONEY, DOROTHEA W. BARTHEL · PubMed · 1965 · 15,488 citations
- The Timed “Up & Go”: A Test of Basic Functional Mobility for Frail Elderly PersonsDiane Podsiadlo, Sandra Richardson · Journal of the American Geriatrics Society · 1991 · 15,001 citations
- National Committee for Clinical Laboratory StandardsErika Bruck · PEDIATRICS · 1980 · 14,735 citations
Questions about this paper
How do I cite this paper in APA style?
Cook, J., Wyatt, M., Wei, P., Chin, I., Gupta, S., van Vuuren, V. Z., Siburian, R., Spicer, A., Viviano, K., Cami, A., Malhotra, R., Yao, Z., Rasley, J., & Kaushik, G. (2026). Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2603.23515
How many times has this paper been cited?
1 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
John Cook, Michael Wyatt, Peng Wei, Iris Chin, Santosh Gupta, Van Zyl van Vuuren, Richie Siburian, Amanda Spicer, Kristen Viviano, Alda Cami, Raunaq Malhotra, Zhewei Yao and 2 more.
What is the DOI of this paper?
10.48550/arxiv.2603.23515, permanently at https://doi.org/10.48550/arxiv.2603.23515. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.