arXiv 2605.250202026-05-24cs.AIcs.CL

慢性皮肤病纵向数据检索中的隐私保护本地语言模型：在天疱疮患者中的实施

Privacy-Preserving Local Language Models for Longitudinal Data Retrieval in Chronic Dermatologic Disease: Implementation in Pemphigus Patients

Division of Systems Medicine, Department of Metabolism, Digestion and Reproduction, Imperial College London（系统医学系，代谢、消化与生殖部，帝国理工学院伦敦分校）；Department of Dermatology and Venereology, Istanbul Research and Training Hospital（皮肤科与性病科，伊斯坦布尔研究与培训医院）；Department of Dermatology and Venereology, Istanbul Medeniyet University（皮肤科与性病科，伊斯坦布尔梅德尼yet大学）；Department of Dermatology and Venereology, Istanbul Medicana Atakoy Hospital（皮肤科与性病科，伊斯坦布尔Medicana阿塔科伊医院）

Abdurrahim Yilmaz, Ayşe Esra Koku Aksu, Duygu Yamen, Vefa Asli Erdemir, Mehmet Salih Gurel, Gulsum Gencoglan, Joram M. Posma, Burak Temelkuran

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究评估了本地部署的隐私保护小型语言模型（SLM）在天疱疮患者长期随访记录中检索结构化临床特征并生成纵向摘要的能力，结果显示SLM在特征检索任务中平均准确率达82.25%，且医生对AI生成摘要的质量、临床准确性和实用性评分较高。

摘要：

慢性皮肤病如天疱疮需要长期随访，产生大量纵向临床文档，在常规就诊期间难以全面审查，增加了临床医生的工作量以及遗漏关键历史信息的风险。我们评估了本地部署的隐私保护小型语言模型（SLM）是否能够从长期皮肤科随访记录中检索结构化临床特征并生成纵向摘要。在这项回顾性病例系列研究中，30名天疱疮患者贡献了541份就诊记录，汇总为完整的纵向记录（89,336词）；由两位皮肤科专家标注了56个临床相关特征。本地部署的SLM（Qwen3 4B Thinking 2507）对每份完整记录进行查询，以检索56个特征并生成一份最终报告摘要。在1,680个特征检索任务中，平均准确率为82.25%。皮肤科医生对AI生成摘要的整体质量（8.23-8.47）、临床准确性（7.93-8.20）和实用性（8.47-8.50）评分较高，评估者间无显著差异，且在53.3%的评估中总体偏好AI摘要。这些发现表明，隐私保护的本地部署SLM可以优于医学专家，并可靠地生成有临床意义的纵向摘要。在适当监督下，SLM可以支持临床决策。

英文摘要：

Chronic dermatologic diseases such as pemphigus require long-term follow-up, generating extensive longitudinal clinical documentation that is difficult to review comprehensively during routine visits and increasing clinician workload as well as the risk of missing critical historical information. We evaluated whether a locally deployed, privacy-preserving small language model (SLM) could retrieve structured clinical features and generate longitudinal summaries from long-term dermatology follow-up records. In this retrospective case series, thirty pemphigus patients contributed 541 visit notes that were aggregated into full longitudinal records (89,336 words); 56 clinically relevant features were annotated by two expert dermatologists. The locally deployed SLM (Qwen3 4B Thinking 2507) was queried with each complete record to retrieve 56 features and generate one final report summaries. Across 1,680 feature retrieval tasks, mean accuracy was 82.25%. Dermatologists' ratings of AI-generated summaries were high for overall quality (8.23-8.47), clinical accuracy (7.93-8.20), and usefulness (8.47-8.50), with no significant inter-evaluator differences and an overall preference for AI summaries in 53.3% of evaluations. These findings suggest that privacy-preserving, locally deployed SLMs can outperform medical experts and reliably generate clinically meaningful longitudinal summaries. SLMs may support clinical decision-making when integrated with appropriate oversight.

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。