跳到正文
arXiv cs.CL· Daniel Reichenpfader (Institute for Patient-Centered Digital Health, Bern University of Applied Sciences, Bern, Switzerland, Faculty of Medicine, University of Geneva, Geneva, Switzerland), Kwaku Appiagyei (F. Hoffmann-La Roche AG, Basel, Switzerland), Manovah Nadarajah (Kantonsspital Aarau, Aarau, Switzerland), Ahmed Reda Bahr (University of Alexandria, Alexandria, Egypt), Irina Filchenko (Department of Neurology, Inselspital, Bern University Hospital and University of Bern, Bern, Switzerland), Fabio Dennst\"adt (Data Semantics and Decision Making Institute, Department of Digital Medicine, Inselspital, Bern University Hospital and University of Bern, Bern, Switzerland)·· 2 天前AI 评分31

LLM 智能体从肉瘤病例报告摘要中自动归纳通用数据元素并抽取信息

Agentic induction of common data elements and information extraction from sarcoma case report abstracts

AI 导读

研究用大语言模型智能体从 24,588 篇 PubMed 摘要中提出 72,909 个候选通用数据元素(CDE),经智能体整合后得到 1,570 个,并保留至少 50 篇摘要支持的 53 个用于全语料证据关联抽取。

正文 · 原文

Authors:Daniel Reichenpfader (1 and 2), Kwaku Appiagyei (3), Manovah Nadarajah (4), Ahmed Reda Bahr (5), Irina Filchenko (6), Fabio Dennstädt (7, 8 and 9) ((1) Institute for Patient-Centered Digital Health, Bern University of Applied Sciences, Bern, Switzerland, (2) Faculty of Medicine, University of Geneva, Geneva, Switzerland, (3) F. Hoffmann-La Roche AG, Basel, Switzerland, (4) Kantonsspital Aarau, Aarau, Switzerland, (5) University of Alexandria, Alexandria, Egypt, (6) Department of Neurology, Inselspital, Bern University Hospital and University of Bern, Bern, Switzerland, (7) Data Semantics and Decision Making Institute (DSDMI), St. Gallen, Switzerland, (8) Department of Radiation Oncology, Inselspital, Bern University Hospital and University of Bern, Bern, Switzerland, (9) Department of Digital Medicine, Inselspital, Bern University Hospital and University of Bern, Bern, Switzerland)

View PDF HTML (experimental)

Abstract:Common data elements (CDEs) support clinical data comparability but are usually defined manually. We investigated whether large language model agents could derive a CDE schema from sarcoma case report abstracts and apply it at scale. An agent proposed 72,909 candidate CDEs from 24,588 PubMed abstracts; agentic consolidation yielded 1,570 CDEs. We retained 53 supported by at least 50 abstracts and attempted evidence-linked extraction across the corpus. Processing succeeded for all retained CDEs in 79.2% of abstracts, including responses indicating non-mention. Two human reviewers approved 83.0% of retained CDEs for relevance, 67.9% for clarity, and 56.6% for correctness. In 100 sampled abstracts, exact agreement with the value extracted by the LLM was 94.9% among 1,116 pairs with a non-empty answer (abstract-level bootstrap 95% CI, 93.0-96.4%). Errors concentrated in anatomical and temporal fields. These findings support agent-assisted schema drafting followed by expert curation before large-scale extraction.
Subjects: Computation and Language (cs.CL)
ACM classes: I.2.7; J.3; H.3.1
Cite as: arXiv:2611.18506 [cs.CL]
  (or arXiv:2611.18506v1 [cs.CL] for this version)
  https://doi.org/10.48550/arXiv.2611.18506

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Daniel Reichenpfader [view email]
[v1] Fri, 13 Nov 2026 18:23:11 UTC (1,603 KB)

来源:arXiv cs.CL · arxiv.org