← Evidence map

ArticleClinical and experimental medicine2026

Benchmarking LLM decision support in inflammatory aneurysms: DeepSeek-R1, DeepSeek-V3 and ChatGPT-4o.

Jianqiang Hao et al.PubMed ↗Full text ↗Publisher ↗

No numbers read from the abstract.

Not cited yet

Full record →Abstract, authors, funding and every citing paper · PMID 41931161