Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Evaluating DeepSeek in Real-World Clinical Scenarios on Performance and Safeguards: Multicenter Comparative Study.

Created on 14 Nov 2026

Authors

Zuyi Yang, Dianzhe Tian, Meiqian He, Erqian Wang, Lvyuxing Zhao, Xinyu Zhao, Xiaokai Chen, Wei Xia, Wei Zhou, Yanhao Ma, Chuanyong Ma, Haiyan Chen, Huiqin Lu, Shulin Liu, Xin Zhao, Yiyao Xu, Xin Lu, Fengdan Wang, Lei Zhang, Youxin Chen

Published in

Journal of medical Internet research. Volume 28. Pages e86746. Nov 13, 2026. Epub Nov 13, 2026.

Abstract

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.
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.
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.
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).
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.

PMID:
43024857
Bibliographic data and abstract were imported from PubMed on 14 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 11
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement