Evaluating the efficacy of using large language models in preoperative prediction of microvascular invasion in HCC: a multicenter study.
運用大型語言模型於肝細胞癌(HCC)術前微血管侵犯預測之效能評估:多中心研究
Sci Rep 2025-07-29
Assessing Artificial Intelligence-Generated Patient Educational Material on Gestational Diabetes Mellitus: Content and Quality Evaluation.
人工智慧生成妊娠糖尿病患者衛教資料之內容與品質評估
J Perinat Neonatal Nurs 2025-07-29
這項研究發現,ChatGPT 和 Gemini 產生的妊娠糖尿病衛教資料,內容易讀、實用,專家評價也很高。雖然 AI 在病人衛教上很有潛力,但還需要更多研究來證實長期效果。
相關文章PubMedDOI推理
A dual-dimension collaborative enhancement framework to boost language model spatial semantic understanding.
提升語言模型空間語意理解的雙維度協同增強架構
Ann N Y Acad Sci 2025-07-29
AI or nay? Evaluating the potential use of ChatGPT (Open AI) and Perplexity AI in undergraduate nursing research: An exploratory case study.
AI 還是不要?評估 ChatGPT(Open AI)與 Perplexity AI 在護理學系大學生研究中的潛在應用:一項探索性個案研究
Nurse Educ Pract 2025-07-29
A Randomized Controlled Trial of a Deep Language Learning Model-Based Simulation Tool for Undergraduate Medical Students in Surgery.
以深度語言學習模型為基礎之模擬工具應用於外科醫學生的隨機對照試驗
J Surg Educ 2025-07-29
Leveraging open-source large language models (LLMs) in scoping reviews: a case study on disability and AI applications.
在範疇性回顧中運用開源大型語言模型(LLMs):以身心障礙與人工智慧應用為案例研究
Int J Med Inform 2025-07-29