Generative pretrained transformer models can function as highly reliable second screeners of titles and abstracts in systematic reviews: A proof of concept and common guidelines.
生成式預訓練轉換器(Generative pretrained transformer, GPT)模型可作為系統性回顧中標題與摘要的高可靠度第二篩選者:概念驗證與通用指引
Psychol Methods 2025-07-10
Large Language Models in Intracardiac Electrogram Interpretation: A New Frontier in Cardiac Diagnostics for Pacemaker Patients.
大型語言模型於心臟起搏器患者心內心電圖判讀:心臟診斷的新前沿
Anatol J Cardiol 2025-07-10
Data Extraction and Curation from Radiology Reports for Pancreatic Cyst Surveillance Using Large Language Models.
利用大型語言模型從放射科報告中擷取與整理胰臟囊腫監測資料
J Am Coll Surg 2025-07-10
Hybrid AI Framework for the Early Detection of Heart Failure: Integrating Traditional Machine Learning and Generative Language Models With Clinical Data.
結合傳統機器學習與生成式語言模型及臨床資料之心臟衰竭早期偵測混合式AI架構
Cureus 2025-07-10
Evaluation and comparison of large language models' responses to questions related optic neuritis.
大型語言模型對視神經炎相關問題回應的評估與比較
Front Med (Lausanne) 2025-07-10
這項研究比較四款熱門聊天機器人在回答視神經炎問題時的表現,發現 ChatGPT-4.0 和 Google Bard 的正確性明顯較佳,尤其 ChatGPT-4.0 整體表現最好。不過,所有機器人的回答都需要大學程度的閱讀能力,臨床應用前還需進一步改進和審慎評估。
相關文章PubMedDOI推理