Optimizing data extraction: harnessing RAG and LLMs for German medical documents

In the field of medical data analysis, converting unstructured text documents into a structured format suitable for further use is a significant challenge. This study introduces an automated local deployed data privacy secure pipeline that uses open-source Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) architecture to convert medical German language documents with sensitive health-related information into a structured format. Testing on a proprietary dataset of 800 unstructured original medical reports demonstrated an accuracy of up to 90% in data extraction of the pipeline compared to data extracted manually by physicians and medical students. This highlights the pipeline’s potential as a valuable tool for efficiently extracting relevant data from unstructured sources.

Metadaten
Author:	Yingding Wang, Simon Leutner, Michael Ingrisch, Christoph Klein, Ludwig Christian Hinske ORCiD GND, Katharina Danhauser
URN:	urn:nbn:de:bvb:384-opus4-1196449
Frontdoor URL	https://opus.bibliothek.uni-augsburg.de/opus4/119644
ISBN:	9781643685335OPAC
ISSN:	0926-9630OPAC
ISSN:	1879-8365OPAC
Parent Title (English):	Digital health and informatics innovations for sustainable health care systems: proceedings of MIE 2024
Publisher:	IOS Press
Place of publication:	Amsterdam
Editor:	John Mantas, Arie Hasman, George Demiris, Kaija Saranto, Michael Marschollek, Theodoros N. Arvanitis, Ivana Ognjanović, Arriel Benis, Parisis Gallos, Emmanouil Zoulias, Elisavet Andrikopoulou
Type:	Conference Proceeding
Language:	English
Year of first Publication:	2024
Publishing Institution:	Universität Augsburg
Release Date:	2025/03/18
First Page:	949
Last Page:	950
Series:	Studies in Health Technology and Informatics ; 316
DOI:	https://doi.org/10.3233/shti240567
Institutes:	Medizinische Fakultät
	Medizinische Fakultät / Universitätsklinikum
	Medizinische Fakultät / Lehrstuhl für Datenmanagement und Clinical Decision Support
Dewey Decimal Classification:	0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 004 Datenverarbeitung; Informatik
	6 Technik, Medizin, angewandte Wissenschaften / 61 Medizin und Gesundheit / 610 Medizin und Gesundheit
Licence (German):	CC-BY-NC 4.0: Creative Commons: Namensnennung - Nicht kommerziell (mit Print on Demand)

Open Access