基于GLINER的局部混淆实现公正的上下文感知溯源：PII移除系统的开发与评估
Local Obfuscation by GLINER for Impartial Context Aware Lineage: Development and evaluation of PII Removal system
AI总结：
针对临床笔记PII移除需求，本研究开发了基于微调GLiNER的本地可部署系统LOGICAL，在精神病院临床数据上取得0.980的微平均F1值，性能优于多款主流方案，可在普通笔记本高效运行，为资源受限环境下的临床数据脱敏提供了实用方案。
摘要：
从电子健康记录（EHR）的临床笔记中移除个人身份信息（PII）对研究与AI开发至关重要。尽管大语言模型（LLM）功能强大，但其高昂的计算成本以及基于API的服务带来的数据隐私风险限制了其应用，在资源匮乏的环境中尤为明显。为解决这一问题，我们开发了LOGICAL（基于GLINER的局部混淆实现公正的上下文感知溯源），这是一款高效、可本地部署的PII移除系统，基于微调的通用轻量命名实体识别（GLiNER）模型构建。\n我们使用了来自某精神病院EHR系统的1515份临床文档，定义了9类需要移除的PII类别。我们在2849个文本样本上微调了modern-gliner-bi-large-v1.0模型，并在包含376个样本的测试集上使用字符级精确率、召回率和F1值进行评估。我们将其性能与Microsoft Azure NER、Microsoft Presidio，以及使用Gemini-Pro-2.5和Llama-3.3-70B-Instruct的零样本提示方法进行了对比。\n微调后的GLiNER模型性能更优，整体微平均F1值达到0.980，显著优于Gemini-Pro-2.5（F1值：0.845）。LOGICAL可完全正确脱敏95%的文档，而次优方案的这一比例仅为64%。该模型可在无专用GPU的标准笔记本电脑上高效运行。不过，2%的实体级假阴性率表明，所有受测试系统都需要人工在环验证。\n像GLiNER这样经过微调的专用Transformer模型为临床笔记的PII移除提供了准确、计算高效且安全的解决方案。这种“源头脱敏”方法是资源密集型LLM的实用替代方案，能够在保护数据隐私的同时，为研究和AI开发构建去标识化数据集，尤其适用于资源受限的环境。
英文摘要：
Removing Personally Identifiable Information (PII) from clinical notes in Electronic Health Records (EHRs) is essential for research and AI development. While Large Language Models (LLMs) are powerful, their high computational costs and the data privacy risks of API-based services limit their use, especially in low-resource settings. To address this, we developed LOGICAL (Local Obfuscation by GLINER for Impartial Context-Aware Lineage), an efficient, locally deployable PII removal system built on a fine-tuned Generalist and Lightweight Named Entity Recognition (GLiNER) model. We used 1515 clinical documents from a psychiatric hospital's EHR system. We defined nine PII categories for removal. A modern-gliner-bi-large-v1.0 model was fine-tuned on 2849 text instances and evaluated on a test set of 376 instances using character-level precision, recall, and F1-score. We compared its performance against Microsoft Azure NER, Microsoft Presidio, and zero-shot prompting with Gemini-Pro-2.5 and Llama-3.3-70B-Instruct. The fine-tuned GLiNER model achieved superior performance, with an overall micro-average F1-score of 0.980, significantly outperforming Gemini-Pro-2.5 (F1-score: 0.845). LOGICAL correctly sanitised 95% of documents completely, compared to 64% for the next-best solution. The model operated efficiently on a standard laptop without a dedicated GPU. However, a 2% entity-level false negative rate underscores the need for human-in-the-loop validation across all tested systems. Fine-tuned, specialised transformer models like GLiNER offer an accurate, computationally efficient, and secure solution for PII removal from clinical notes. This "sanitisation at the source" approach is a practical alternative to resource-intensive LLMs, enabling the creation of de-identified datasets for research and AI development while preserving data privacy, particularly in resource-constrained environments.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
