Journal article · 2026 · Open access

Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance

Guangyao Zheng, Peter I. Kamel, Jay J. Pillai, Alireza Akhbardeh, Vladimir Braverman, Michael A. Jacobs, Vishwa S. Parekh

Journal of Imaging Informatics in Medicine · Published 13 March 2026 · DOI 10.1007/s10278-026-01912-4

Open PDFView at the publisherCite this paper

In brief

“Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance” is a 2026 journal article by Guangyao Zheng and 6 co-authors, published in Journal of Imaging Informatics in Medicine. OpenAlex counts 3 citations as of 19 November 2026, which puts it in the top 4% of works published in 2026 by citations. It is open access under a CC BY-NC-ND licence: the full text is free to read at Journal of Imaging Informatics in Medicine.

Citations
3
Most cited in
2026 (3)
Field-weighted impact
3.08
References
29
Authors
7

Abstract

The purpose of this study is to evaluate whether a single, fine-tuned SLM can match or exceed the performance of LLMs across diverse clinical tasks, enabling hospitals to build tailored, privacy-preserving, efficient, and deployable language models that do not require managing multiple task-specific systems. We used SLMs of varying sizes and applied low-rank adaptation (LoRA) for fine-tuning across three clinical tasks: (1) medical report labeling, (2) DICOM series description harmonization, and (3) impression generation from findings. These tasks were constructed using two datasets: the public Open-i Indiana University Chest X-ray Dataset and an in-house brain MRI DICOM metadata dataset. We compared single-task SLMs, a multi-task SLM (representing our proposed configuration), and GPT-4o using zero-shot and few-shot prompting. We found OPT-350 m to be the optimal SLM. In medical report labeling, the multi-task SLM achieved an F1 score of 0.894 compared to additional prompt-engineered GPT-4o's 0.728. In DICOM series description harmonization, the multi-task achieved an accuracy of 0.975 compared to additional prompt-engineered GPT-4o's 0.878. In impression generation from findings, the multi-task SLM achieved an average Likert scale score of 4.39 ± 1.00, compared to GPT-4o's 3.65 ± 1.00 (p = 0.0008). This study demonstrates that a single fine-tuned SLM can serve as a general-purpose clinical assistant, offering performance on par with or better than larger models. With lower resource requirements, greater customizability, privacy protection, and strong task generalization, fine-tuning one SLM to support multiple clinical tasks meets the practical demands of clinical AI deployment in both high-resource and resource-limited healthcare settings.

Cite this paper

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

APA 7

Zheng, G., Kamel, P. I., Pillai, J. J., Akhbardeh, A., Braverman, V., Jacobs, M. A., & Parekh, V. S. (2026). Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance. Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-026-01912-4

MLA 9

Zheng, Guangyao, et al. “Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance.” Journal of Imaging Informatics in Medicine, 2026. https://doi.org/10.1007/s10278-026-01912-4.

Chicago

Zheng, Guangyao, Peter I. Kamel, Jay J. Pillai, Alireza Akhbardeh, Vladimir Braverman, Michael A. Jacobs, and Vishwa S. Parekh. 2026. “Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance.” Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-026-01912-4.

Harvard

Zheng, G. et al. (2026) ‘Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance’, Journal of Imaging Informatics in Medicine. doi:10.1007/s10278-026-01912-4.

IEEE

G. Zheng et al., “Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance,” Journal of Imaging Informatics in Medicine, 2026, doi: 10.1007/s10278-026-01912-4.

BibTeX

@article{zheng2026adaptive,
  title     = {{Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance}},
  author    = {Zheng, Guangyao and Kamel, Peter I. and Pillai, Jay J. and Akhbardeh, Alireza and Braverman, Vladimir and Jacobs, Michael A. and Parekh, Vishwa S.},
  journal   = {Journal of Imaging Informatics in Medicine},
  year      = {2026},
  doi       = {10.1007/s10278-026-01912-4},
  url       = {https://doi.org/10.1007/s10278-026-01912-4}
}

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

Topic and keywords

Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education

  • small language models
  • multi-task learning
  • low-rank adaptation
  • chest radiography
  • brain MRI

Work with this paper

Most-cited papers in Medicine

All highly cited Medicine papers

Questions about this paper

How do I cite this paper in APA style?

Zheng, G., Kamel, P. I., Pillai, J. J., Akhbardeh, A., Braverman, V., Jacobs, M. A., & Parekh, V. S. (2026). Adaptive, Privacy-Preserving Small Language Models for Multi-Task Clinical Assistance. Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-026-01912-4

How many times has this paper been cited?

3 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 Journal of Imaging Informatics in Medicine: sign in and open the PDF from this page.

Who wrote this paper?

Guangyao Zheng, Peter I. Kamel, Jay J. Pillai, Alireza Akhbardeh, Vladimir Braverman, Michael A. Jacobs, Vishwa S. Parekh.

What is the DOI of this paper?

10.1007/s10278-026-01912-4, permanently at https://doi.org/10.1007/s10278-026-01912-4. 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.