Evaluating the utility of ChatGPT in addressing conceptual and non-conceptual questions related to urodynamic quality control and trace analysis.
ChatGPT 在解決與尿動力學品質控制及圖譜分析相關的概念性與非概念性問題之應用評估
Sci Rep 2025-06-24
A Dataset of Medical Questions Paired with Automatically Generated Answers and Evidence-supported References.
一個包含醫學問題、由自動生成答案及有證據支持參考文獻配對的資料集
Sci Data 2025-06-24
MedAESQA 資料集專為提升大型語言模型在醫療問答的表現而設計,內含 40 個真實醫療問題,每題有 30 答案(人類與 AI 產生),並附上相關科學文獻摘要。每個答案都經人工評估正確性與來源相關性,目的是讓 AI 回答更貼近事實、具科學根據。
相關文章PubMedDOI推理
Artificial Intelligence Language Models to Translate Professional Radiology Mammography Reports Into Plain Language - Impact on Interpretability and Perception by Patients.
人工智慧語言模型將專業放射線乳房攝影報告翻譯為淺顯語言——對病患可理解性與感受的影響
Acad Radiol 2025-06-23
這項研究發現,ChatGPT-4 和 ChatGPT-4o 翻譯乳房攝影和超音波報告,比 Google Gemini 更能讓病患看得懂、覺得貼心。年紀較大或沒做過乳房攝影的人,對AI報告評價更高;教育程度高的人則較挑剔。大多數人都覺得AI翻譯比傳統醫療用語好,顯示AI有助提升病患理解和醫病溝通。
相關文章PubMedDOI推理
Generative AI models: the next anaesthetic agent?
生成式 AI 模型:下一代麻醉劑?
Br J Anaesth 2025-06-23
MacKay 團隊用大型語言模型,能從自由書寫的心臟超音波報告中,自動抓出結構化資料,大幅減少人工處理和錯誤。這展現 AI 在心臟麻醉和手術照護資料處理上的效率與準確性提升。
相關文章PubMedDOI推理
Large Language Model-Assisted Surgical Consent Forms in Non-English Language: Content Analysis and Readability Evaluation.
大型語言模型輔助的非英語手術同意書:內容分析與可讀性評估
J Med Internet Res 2025-06-19
Bridging Health Literacy Gaps in Spine Care: Using ChatGPT-4o to Improve Patient-Education Materials.
彌補脊椎照護健康素養差距:運用 ChatGPT-4o 改善病患衛教資料
J Bone Joint Surg Am 2025-06-19
Sentiment Analysis Using a Large Language Model-Based Approach to Detect Opioids Mixed With Other Substances Via Social Media: Method Development and Validation.
利用大型語言模型進行情感分析,透過社群媒體偵測與其他物質混合的鴉片類藥物:方法開發與驗證
JMIR Infodemiology 2025-06-19
Data extraction from free-text stroke CT reports using GPT-4o and Llama-3.3-70B: the impact of annotation guidelines.
使用 GPT-4o 與 Llama-3.3-70B 從自由文本中風中風電腦斷層報告擷取資料:標註指引的影響
Eur Radiol Exp 2025-06-22