Metraipolis › archive › dc_73ef57d97e73

From PDF to Evidence: Structure-Aware Retrieval for Clinical Practice Guidelines — arXiv:2609.33447

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33447
License (as archived)
CC0-1.0
Archived
2026-10-07T23:49:49.509Z
SHA-256 of the archived text
92ba1ddee09591ca274593fc584e51c68586454bb947b8ff546ef72f674aa49d
Stable URL of this page
https://metraipolis.com/documento/dc_73ef57d97e73

How to cite

From PDF to Evidence: Structure-Aware Retrieval for Clinical Practice Guidelines — arXiv:2609.33447 — arXiv (en), https://arxiv.org/abs/2609.33447 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_73ef57d97e73 (Metraipolis archive dc_73ef57d97e73, sha256 92ba1ddee09591ca274593fc584e51c68586454bb947b8ff546ef72f674aa49d)
BibTeX
@misc{dc73ef57d97e73,
  title = {From PDF to Evidence: Structure-Aware Retrieval for Clinical Practice Guidelines — arXiv:2609.33447},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33447},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_73ef57d97e73; sha256 92ba1ddee09591ca274593fc584e51c68586454bb947b8ff546ef72f674aa49d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33447. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Guideline documents are published as unstructured PDFs whose evidence is locked in visual structures---tables, flowcharts, and graded recommendations---that standard retrieval pipelines flatten into fixed-size text chunks. We cast evidence access as a document image analysis problem: parse each page image into typed structural elements, then retrieve structure-aware evidence units that follow the document's own layout (sections, table rows, flowchart paths, graded recommendations), each keeping its structural context so a result points to a specific element rather than a page. On 26 clinical practice guidelines from 9 sources (3,619 pages, Chinese and English) with 199 evidence queries, structure-aware units rank the gold element first under BM25, dense, and hybrid retrieval (hybrid Element Hit@1 of 0.382), with a significant element-level ranking gain over per-element OCR text (MRR_e +0.107, p=0.002; the Hit@5 gain is directional, p=0.17), while matching page-level recall (Page Hit@5 0.879 vs. 0.889, p=0.75) at 3.8x less context and clearly outperforming a ColPali visual-RAG baseline (PH@5 0.497).

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.IR).
Categorie arXiv: Information Retrieval (cs.IR) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.33447 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T23:26:57.191Z dalla pagina https://arxiv.org/catchup/cs.IR/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).