# From PDF to Evidence: Structure-Aware Retrieval for Clinical Practice Guidelines

arXiv: https://arxiv.org/abs/2609.33447 · HTML: https://mlsanity.com/paper/2609.33447/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 29 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Builds on (most cited first, 23 of 23)

- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [PubLayNet: largest dataset ever for document layout analysis](https://mlsanity.com/paper/1908.07836/) (2019)
- [Benchmarking Retrieval-Augmented Generation for Medicine](https://mlsanity.com/paper/2402.13178/) (2024)
- [ColPali: Efficient Document Retrieval with Vision Language Models](https://mlsanity.com/paper/2407.01449/) (2024)
- [DeepSeek-OCR: Contexts Optical Compression](https://mlsanity.com/paper/2510.18234/) (2025)
- [OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotations](https://mlsanity.com/paper/2412.07626/) (2024)
- [DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis](https://mlsanity.com/paper/2206.01062/) (2022)
- [Almanac: Retrieval-Augmented Language Models for Clinical Medicine](https://mlsanity.com/paper/2303.01229/) (2023)
- [TableFormer: Table Structure Understanding with Transformers](https://mlsanity.com/paper/2203.01017/) (2022)
- [dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model](https://mlsanity.com/paper/2512.02498/) (2025)
- [Improving Medical Reasoning through Retrieval and Self-Reflection with Retrieval-Augmented Large Language Models](https://mlsanity.com/paper/2401.15269/) (2024)
- [MedDM:LLM-executable clinical guidance tree for clinical decision-making](https://mlsanity.com/paper/2312.02441/) (2023)
- [Detect-Order-Construct: A Tree Construction based Approach for Hierarchical Document Structure Analysis](https://mlsanity.com/paper/2401.11874/) (2024)
- [RegionRAG: Region-level Retrieval-Augmented Generation for Visual Document Understanding](https://mlsanity.com/paper/2510.27261/) (2025)
- [Grounding Large Language Models in Clinical Evidence: A Retrieval-Augmented Generation System for Querying UK NICE Clinical Guidelines](https://mlsanity.com/paper/2510.02967/) (2025)
- [Unlimited OCR Works](https://mlsanity.com/paper/2606.23050/) (2026)
- [GGPONC: A Corpus of German Medical Text with Rich Metadata Based on Clinical Practice Guidelines](https://mlsanity.com/paper/2007.06400/) (2020)
- [A Decade-Scale Benchmark Evaluating LLMs' Clinical Practice Guidelines Detection and Adherence in Multi-turn Conversations](https://mlsanity.com/paper/2603.25196/) (2026)
- [Guideline2Graph: Profile-Aware Multimodal Parsing for Executable Clinical Decision Graphs](https://mlsanity.com/paper/2604.02477/) (2026)
- [ClinicBot: A Guideline-Grounded Clinical Chatbot with Prioritized Evidence RAG and Verifiable Citations](https://mlsanity.com/paper/2605.00846/) (2026)
- [From PDF to RAG-Ready: Evaluating Document Conversion Frameworks for Domain-Specific Question Answering](https://mlsanity.com/paper/2604.04948/) (2026)
- [SPIRE: Structure-Preserving Interpretable Retrieval of Evidence](https://mlsanity.com/paper/2604.20849/) (2026)
- [SEM-RAG: Structure-Preserving Multimodal Graph Compilation and Entropy-Guided Retrieval for Telecommunication Standards](https://mlsanity.com/paper/2605.08997/) (2026)

## Abstract (arXiv)

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).