Journal article · 2026 · Open access
Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study
Journal of Medical Internet Research, 28, e86746 · Published 13 November 2026 · DOI 10.2196/86746
In brief
“Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study” is a 2026 journal article by Zuyi Yang and 19 co-authors, published in Journal of Medical Internet Research. OpenAlex counts 0 citations as of 18 November 2026. It is open access under a CC BY licence: the full text is free to read at Journal of Medical Internet Research.
- Citations
- 0
- References
- 37
- Authors
- 20
Abstract
Background: DeepSeek, an open-source large language model (LLM) extensively adopted in hospitals across China, presents a locally deployable alternative to proprietary models such as GPT. However, its real-world clinical performance and safeguards remain underexplored. Objective: The aim of this study is to evaluate the performance of the general-purpose and reasoning versions of DeepSeek (DeepSeek AI) and GPT models in diagnosis, medical record analysis, and safeguards against adversarial vulnerabilities. Methods: This cross-sectional study included 50 real-world medical records from 10 hospitals spanning 14 specialties. A total of 7 LLMs were evaluated, including DeepSeek-V3, DeepSeek-R1, GPT-4o (via ChatGPT; OpenAI), o3-mini (via ChatGPT), o3-mini-high (via ChatGPT), GPT-4o (via Copilot; Microsoft), and GPT-4o-reasoning (via Copilot). Models were tasked with generating diagnoses based on admission notes, answering 12 questions related to extraction, summarization, and inference using discharge summaries, and undergoing 4 rounds of simulated adversarial prompt attacks designed to elicit misinformation. The diagnostic evaluation was performed on 47 records after the exclusion of 3 flawed admission notes, while the other tasks used the full set of 50 records. Diagnostic performance was evaluated based on accuracy, with the causes of misdiagnosis further categorized. Medical record analysis was assessed by scores in extraction, summarization, and inference tasks. Safeguard evaluation included identifying the earliest round of successful prompt attack and verifying reference authenticity and accuracy. Results: Among the 7 models included in this study, DeepSeek-R1 and GPT-4o-reasoning (via Copilot) achieved the highest diagnostic accuracy at 64% (30/47). In medical record analysis, both models also excelled across the 3 subdimensions. GPT-4o-reasoning (via Copilot) achieved accuracies of 97.7% (293/300) in extraction, 87.5% (175/200) in summarization, and 87% in inference, while DeepSeek-R1 reached 97.7% (293/300), 83.5% (167/200), and 85% (85/100), respectively. In simulated adversarial safeguard evaluation involving 4 rounds of prompt attacks, all evaluated models showed high vulnerability. DeepSeek-R1 generated medical disinformation in 98% (49/50) of cases, with only a 2% refusal rate. GPT-4o-reasoning (via Copilot) showed the highest refusal rate at 30%, but it still generated disinformation in 70% (35/50) of cases. True hallucination rates ranged from 38.6% (27/70; GPT-4o-reasoning via Copilot) to 73% (73/100; DeepSeek-V3). Conclusions: These findings underscore the clinical application potential of DeepSeek models, while also emphasizing the critical need to strengthen safeguards against adversarial vulnerabilities prior to their clinical deployment.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Yang, Z., Tian, D., He, M., Wang, E., Zhao, L., Zhao, X., Chen, X., Xia, W. J., Zhou, W., Ma, Y., Ma, C., Chen, H., Lu, H., Liu, S. X., Zhao, X., Xu, Y., Lü, X., Wang, F., Zhang, L., & Chen, Y. (2026). Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study. Journal of Medical Internet Research, 28, e86746. https://doi.org/10.2196/86746
MLA 9
Yang, Zuyi, et al. “Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study.” Journal of Medical Internet Research, vol. 28, 2026, p. e86746. https://doi.org/10.2196/86746.
Chicago
Yang, Zuyi, Dianzhe Tian, Meiqian He, Erqian Wang, L. Zhao, Xinyu Zhao, Xiaokai Chen, et al. 2026. “Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study.” Journal of Medical Internet Research 28: e86746. https://doi.org/10.2196/86746.
Harvard
Yang, Z. et al. (2026) ‘Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study’, Journal of Medical Internet Research, 28, p. e86746. doi:10.2196/86746.
IEEE
Z. Yang et al., “Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study,” Journal of Medical Internet Research, vol. 28, p. e86746, 2026, doi: 10.2196/86746.
BibTeX
@article{yang2026evaluating,
title = {{Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study}},
author = {Yang, Zuyi and Tian, Dianzhe and He, Meiqian and Wang, Erqian and Zhao, L. and Zhao, Xinyu and Chen, Xiaokai and Xia, Wei J. and Zhou, Wei and Ma, Yanhao and Ma, Chuanyong and Chen, Haiyan and Lu, Huiqin and Liu, S X and Zhao, Xin and Xu, Yiyao and Lü, Xin and Wang, Fengdan and Zhang, Lei and Chen, Yuehua},
journal = {Journal of Medical Internet Research},
year = {2026},
volume = {28},
pages = {e86746},
doi = {10.2196/86746},
url = {https://doi.org/10.2196/86746}
}
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
- DeepSeek
- large language models
- diagnosis
- medical records review
- misinformation
- GPT-4o
- DeepSeek-R1
- DeepSeek-V3
Work with this paper
Most-cited papers in Medicine
- Yaşlı Kadınlarda Üreme SağlığıDemet Kokanalı, Yaprak Engin Üstün · Jinekoloji-Obstetrik ve Neonatoloji Tıp Dergisi · 2019 · 169,630 citations
- Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 CountriesHyuna Sung, Adalberto M. Filho, Mathieu Laversanne and 5 more · CA A Cancer Journal for Clinicians · 2021 · 119,885 citations
- “Mini-mental state”Marshal F. Folstein, Susan E. Folstein, Paul R. McHugh · Journal of Psychiatric Research · 1975 · 90,424 citations
- Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countriesFreddie Bray, Jacques Ferlay, Isabelle Soerjomataram and 3 more · CA A Cancer Journal for Clinicians · 2018 · 87,396 citations
- Hallmarks of Cancer: The Next GenerationDouglas Hanahan, Robert A. Weinberg · Cell · 2011 · 69,758 citations
Questions about this paper
How do I cite this paper in APA style?
Yang, Z., Tian, D., He, M., Wang, E., Zhao, L., Zhao, X., Chen, X., Xia, W. J., Zhou, W., Ma, Y., Ma, C., Chen, H., Lu, H., Liu, S. X., Zhao, X., Xu, Y., Lü, X., Wang, F., Zhang, L., & Chen, Y. (2026). Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study. Journal of Medical Internet Research, 28, e86746. https://doi.org/10.2196/86746
How many times has this paper been cited?
0 times, according to OpenAlex on 18 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 Medical Internet Research: sign in and open the full text from this page.
Who wrote this paper?
Zuyi Yang, Dianzhe Tian, Meiqian He, Erqian Wang, L. Zhao, Xinyu Zhao, Xiaokai Chen, Wei J. Xia, Wei Zhou, Yanhao Ma, Chuanyong Ma, Haiyan Chen and 8 more.
What is the DOI of this paper?
10.2196/86746, permanently at https://doi.org/10.2196/86746. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 18 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.