Sandmann Sarah, Hegselmann Stefan, Fujarski Michael, Bickmann Lucas, Wild Benjamin, Eils Roland, Varghese Julian
Institute of Medical Informatics, University of Münster, Münster, Germany.
Center for Digital Health, Berlin Institute of Health, Charité - University Medicine Berlin, Berlin, Germany.
Nat Med. 2025 Apr 23. doi: 10.1038/s41591-025-03727-2.
Large language models (LLMs) are increasingly transforming medical applications. However, proprietary models such as GPT-4o face significant barriers to clinical adoption because they cannot be deployed on site within healthcare institutions, making them noncompliant with stringent privacy regulations. Recent advancements in open-source LLMs such as DeepSeek models offer a promising alternative because they allow efficient fine-tuning on local data in hospitals with advanced information technology infrastructure. Here, to demonstrate the clinical utility of DeepSeek-V3 and DeepSeek-R1, we benchmarked their performance on clinical decision support tasks against proprietary LLMs, including GPT-4o and Gemini-2.0 Flash Thinking Experimental. Using 125 patient cases with sufficient statistical power, covering a broad range of frequent and rare diseases, we found that DeepSeek models perform equally well and in some cases better than proprietary LLMs. Our study demonstrates that open-source LLMs can provide a scalable pathway for secure model training enabling real-world medical applications in accordance with data privacy and healthcare regulations.
大语言模型（LLMs）正在日益改变医学应用。然而，像GPT-4o这样的专有模型在临床应用上面临重大障碍，因为它们无法在医疗机构内部署，这使得它们不符合严格的隐私法规。诸如DeepSeek模型等开源大语言模型的最新进展提供了一个有前景的替代方案，因为它们允许在拥有先进信息技术基础设施的医院中对本地数据进行高效微调。在此，为了证明DeepSeek-V3和DeepSeek-R1的临床效用，我们将它们在临床决策支持任务上的性能与专有大语言模型进行了基准测试，包括GPT-4o和Gemini-2.0 Flash Thinking Experimental。使用125个具有足够统计效力的患者病例，涵盖广泛的常见和罕见疾病，我们发现DeepSeek模型表现同样出色，在某些情况下甚至优于专有大语言模型。我们的研究表明，开源大语言模型可以提供一条可扩展的途径，用于安全的模型训练，从而根据数据隐私和医疗法规实现真实世界的医学应用。
