DIRI: Adversarial Patient Reidentification with Large Language Models for Evaluating Clinical Text Anonymization.
DIRI:利用大型語言模型進行對抗性病患再識別以評估臨床文本去識別化
AMIA Jt Summits Transl Sci Proc 2025-06-12
Robust privacy amidst innovation with large language models through a critical assessment of the risks.
在大型語言模型創新中的穩健隱私:對風險的批判性評估。
J Am Med Inform Assoc 2025-03-20
Evaluation and Bias Analysis of Large Language Models in Generating Synthetic Electronic Health Records: Comparative Study.
大型語言模型在生成合成電子健康紀錄的評估與偏誤分析:比較性研究
J Med Internet Res 2025-05-12
Evaluating LLMs' Potential to Identify Rare Patient Identifiers in Patient Health Records.
評估大型語言模型(LLMs)在辨識病歷中罕見病人識別資訊的潛力
Stud Health Technol Inform 2025-05-17
Exposing Vulnerabilities in Clinical LLMs Through Data Poisoning Attacks: Case Study in Breast Cancer.
透過資料投毒攻擊揭露臨床大型語言模型(LLMs)之弱點:以乳癌為案例研究
AMIA Annu Symp Proc 2025-05-26
Ethical Privacy Framework for Large Language Models in Smart Healthcare: A Comprehensive Evaluation and Protection Approach.
智慧醫療中大型語言模型的倫理隱私框架:全面性評估與保護方法
IEEE J Biomed Health Inform 2025-06-04
Harnessing Moderate-Sized Language Models for Reliable Patient Data Deidentification in Emergency Department Records: Algorithm Development, Validation, and Implementation Study.
運用中等規模語言模型於急診部門紀錄中進行可靠的病患資料去識別化:演算法開發、驗證與實作研究
JMIR AI 2025-07-03