Journal article · 2024 · Open access

Privacy-preserving large language models for structured medical information retrieval

Isabella C. Wiest, Dyke Ferber, Jiefu Zhu, Marko van Treeck, Sonja K. Meyer, Radhika Juglan, Zunamys Itzell Carrero, Daniel Paech, Jens Kleesiek, Matthias Ebert, Daniel Truhn, Jakob Nikolas Kather

npj Digital Medicine, 7(1), 257 · Published 20 September 2024 · DOI 10.1038/s41746-024-01233-2

Open PDFView at the publisherCite this paper

In brief

“Privacy-preserving large language models for structured medical information retrieval” is a 2024 journal article by Isabella C. Wiest and 11 co-authors, published in npj Digital Medicine. OpenAlex counts 223 citations as of 19 November 2026, which puts it in the top 1% of works published in 2024 by citations. It is open access under a CC BY licence: the full text is free to read at npj Digital Medicine.

Citations
223
Most cited in
2026 (124)
Field-weighted impact
19.89
References
45
Authors
12

Abstract

Most clinical information is encoded as free text, not accessible for quantitative analysis. This study presents an open-source pipeline using the local large language model (LLM) “Llama 2” to extract quantitative information from clinical text and evaluates its performance in identifying features of decompensated liver cirrhosis. The LLM identified five key clinical features in a zero- and one-shot manner from 500 patient medical histories in the MIMIC IV dataset. We compared LLMs of three sizes and various prompt engineering approaches, with predictions compared against ground truth from three blinded medical experts. Our pipeline achieved high accuracy, detecting liver cirrhosis with 100% sensitivity and 96% specificity. High sensitivities and specificities were also yielded for detecting ascites (95%, 95%), confusion (76%, 94%), abdominal pain (84%, 97%), and shortness of breath (87%, 97%) using the 70 billion parameter model, which outperformed smaller versions. Our study successfully demonstrates the capability of locally deployed LLMs to extract clinical information from free text with low hardware requirements.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Wiest, I. C., Ferber, D., Zhu, J., van Treeck, M., Meyer, S. K., Juglan, R., Carrero, Z. I., Paech, D., Kleesiek, J., Ebert, M., Truhn, D., & Kather, J. N. (2024). Privacy-preserving large language models for structured medical information retrieval. npj Digital Medicine, 7(1), 257. https://doi.org/10.1038/s41746-024-01233-2

MLA 9

Wiest, Isabella C., et al. “Privacy-preserving large language models for structured medical information retrieval.” npj Digital Medicine, vol. 7, no. 1, 2024, p. 257. https://doi.org/10.1038/s41746-024-01233-2.

Chicago

Wiest, Isabella C., Dyke Ferber, Jiefu Zhu, Marko van Treeck, Sonja K. Meyer, Radhika Juglan, Zunamys Itzell Carrero, et al. 2024. “Privacy-preserving large language models for structured medical information retrieval.” npj Digital Medicine 7 (1): 257. https://doi.org/10.1038/s41746-024-01233-2.

Harvard

Wiest, I. C. et al. (2024) ‘Privacy-preserving large language models for structured medical information retrieval’, npj Digital Medicine, 7(1), p. 257. doi:10.1038/s41746-024-01233-2.

IEEE

I. C. Wiest et al., “Privacy-preserving large language models for structured medical information retrieval,” npj Digital Medicine, vol. 7, no. 1, p. 257, 2024, doi: 10.1038/s41746-024-01233-2.

BibTeX

@article{wiest2024privacy,
  title     = {{Privacy-preserving large language models for structured medical information retrieval}},
  author    = {Wiest, Isabella C. and Ferber, Dyke and Zhu, Jiefu and van Treeck, Marko and Meyer, Sonja K. and Juglan, Radhika and Carrero, Zunamys Itzell and Paech, Daniel and Kleesiek, Jens and Ebert, Matthias and Truhn, Daniel and Kather, Jakob Nikolas},
  journal   = {npj Digital Medicine},
  year      = {2024},
  volume    = {7},
  number    = {1},
  pages     = {257},
  doi       = {10.1038/s41746-024-01233-2},
  url       = {https://doi.org/10.1038/s41746-024-01233-2}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
20241818
20258181
2026124124

Its field-weighted citation impact is 19.89: it has been cited 19.89 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare

  • large language models
  • Llama 2
  • clinical text extraction
  • decompensated cirrhosis
  • MIMIC-IV
  • prompt engineering
  • zero-shot learning
  • one-shot learning

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Wiest, I. C., Ferber, D., Zhu, J., van Treeck, M., Meyer, S. K., Juglan, R., Carrero, Z. I., Paech, D., Kleesiek, J., Ebert, M., Truhn, D., & Kather, J. N. (2024). Privacy-preserving large language models for structured medical information retrieval. npj Digital Medicine, 7(1), 257. https://doi.org/10.1038/s41746-024-01233-2

How many times has this paper been cited?

223 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 npj Digital Medicine: sign in and open the PDF from this page.

Who wrote this paper?

Isabella C. Wiest, Dyke Ferber, Jiefu Zhu, Marko van Treeck, Sonja K. Meyer, Radhika Juglan, Zunamys Itzell Carrero, Daniel Paech, Jens Kleesiek, Matthias Ebert, Daniel Truhn, Jakob Nikolas Kather.

What is the DOI of this paper?

10.1038/s41746-024-01233-2, permanently at https://doi.org/10.1038/s41746-024-01233-2. 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.