@inproceedings{cinar-koras-etal-2026-configurable,
title = "Configurable Clinical Information Extraction with Agentic {RAG}: What Works, What Breaks, and Why",
author = "{\c{C}}inar-Kora{\c{s}}, Osman Alperen and
Bauer, Marie and
Khattab, Sameh and
Engelke, Merlin and
Kim, Moon and
Settelmeier, Stephan and
Sugawara, Shigeyasu and
Freisleben, Fabian and
Nensa, Felix and
Kleesiek, Jens",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.50/",
pages = "827--839",
ISBN = "979-8-89176-460-6",
abstract = "Patient contexts span hundreds of heterogeneous documents and thousands of structured data points, yet the document-level metadata that AI systems need for retrieval and triage is absent or incomplete. Standard retrieval-augmented generation fails on this data, mishandling temporal reasoning, cross-document dependencies, and missing metadata. We report on the deployment of ACIE (Agentic Clinical Information Extraction) at University Medicine Essen: an on-premise agentic RAG pipeline that reasons over complete patient contexts and grounds every answer in source passages for clinician verification. We quantify the metadata gap, trace the architectural decisions it shaped, and evaluate extraction alongside an independent retrospective lymphoma registry study, in which nuclear-medicine physicians verify every extracted value against its cited sources. Across 7,326 clinician judgments on this lymphoma-registry workflow, 96.5{\%} of extractions were accepted, with per-type acceptance ranging from 80{\%} to 99{\%}."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cinar-koras-etal-2026-configurable">
<titleInfo>
<title>Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why</title>
</titleInfo>
<name type="personal">
<namePart type="given">Osman</namePart>
<namePart type="given">Alperen</namePart>
<namePart type="family">Çinar-Koraş</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie</namePart>
<namePart type="family">Bauer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sameh</namePart>
<namePart type="family">Khattab</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Merlin</namePart>
<namePart type="family">Engelke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Moon</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stephan</namePart>
<namePart type="family">Settelmeier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shigeyasu</namePart>
<namePart type="family">Sugawara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabian</namePart>
<namePart type="family">Freisleben</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Felix</namePart>
<namePart type="family">Nensa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jens</namePart>
<namePart type="family">Kleesiek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pepa</namePart>
<namePart type="family">Atanasova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gholamreza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-460-6</identifier>
</relatedItem>
<abstract>Patient contexts span hundreds of heterogeneous documents and thousands of structured data points, yet the document-level metadata that AI systems need for retrieval and triage is absent or incomplete. Standard retrieval-augmented generation fails on this data, mishandling temporal reasoning, cross-document dependencies, and missing metadata. We report on the deployment of ACIE (Agentic Clinical Information Extraction) at University Medicine Essen: an on-premise agentic RAG pipeline that reasons over complete patient contexts and grounds every answer in source passages for clinician verification. We quantify the metadata gap, trace the architectural decisions it shaped, and evaluate extraction alongside an independent retrospective lymphoma registry study, in which nuclear-medicine physicians verify every extracted value against its cited sources. Across 7,326 clinician judgments on this lymphoma-registry workflow, 96.5% of extractions were accepted, with per-type acceptance ranging from 80% to 99%.</abstract>
<identifier type="citekey">cinar-koras-etal-2026-configurable</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.50/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>827</start>
<end>839</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why
%A Çinar-Koraş, Osman Alperen
%A Bauer, Marie
%A Khattab, Sameh
%A Engelke, Merlin
%A Kim, Moon
%A Settelmeier, Stephan
%A Sugawara, Shigeyasu
%A Freisleben, Fabian
%A Nensa, Felix
%A Kleesiek, Jens
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F cinar-koras-etal-2026-configurable
%X Patient contexts span hundreds of heterogeneous documents and thousands of structured data points, yet the document-level metadata that AI systems need for retrieval and triage is absent or incomplete. Standard retrieval-augmented generation fails on this data, mishandling temporal reasoning, cross-document dependencies, and missing metadata. We report on the deployment of ACIE (Agentic Clinical Information Extraction) at University Medicine Essen: an on-premise agentic RAG pipeline that reasons over complete patient contexts and grounds every answer in source passages for clinician verification. We quantify the metadata gap, trace the architectural decisions it shaped, and evaluate extraction alongside an independent retrospective lymphoma registry study, in which nuclear-medicine physicians verify every extracted value against its cited sources. Across 7,326 clinician judgments on this lymphoma-registry workflow, 96.5% of extractions were accepted, with per-type acceptance ranging from 80% to 99%.
%U https://aclanthology.org/2026.emnlp-industry.50/
%P 827-839
Markdown (Informal)
[Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why](https://aclanthology.org/2026.emnlp-industry.50/) (Çinar-Koraş et al., EMNLP 2026)
ACL
- Osman Alperen Çinar-Koraş, Marie Bauer, Sameh Khattab, Merlin Engelke, Moon Kim, Stephan Settelmeier, Shigeyasu Sugawara, Fabian Freisleben, Felix Nensa, and Jens Kleesiek. 2026. Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 827–839, Budapest, Hungary. Association for Computational Linguistics.