EUVIMEDEUVIMED entdecken
Crossref

Retrieval-Augmented Language Models for Patient-Centered Periprocedural Anticoagulation in Interventional Radiology

Hossam A Zaki, Allison Brea, Krishnaveni Parvataneni, Hadeel Saab, Christopher P Bridge, Satvik Tripathi, Dania Daye

CardioVascular and Interventional Radiology · 2026

DOI: 10.1007/s00270-026-04428-0 ↗

Kurz gefasst

Fachartikel in der Zeitschrift CardioVascular and Interventional Radiology, 2026. Verfasst von Hossam A Zaki und 6 weiteren Autor:innen. Beteiligte Einrichtung: The Warren Alpert Medical School of Brown University, Providence, RI, USA. hossam_zaki@brown.edu. Hauptthemen laut MEDLINE: Large Language Models, Anticoagulants, Radiology, Interventional. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 16 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: This study evaluates Large Language Models (LLMs) integrated with Retrieval-Augmented Generation (RAG) frameworks for generating accurate, guideline-concordant anticoagulation (AC) recommendations in Interventional Radiology (IR) MATERIALS AND METHODS: 394 IR procedure-AC pairs were extracted from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Five state-of-the-art LLMs, Claude-3.5-Sonnet, GPT-4o, LLaMA-3.1, Mistral-Large-2411, and Qwen-2.5, were deployed within a RAG system built using LlamaIndex, which indexed the Society of Interventional Radiology (SIR) guidelines. Claude was evaluated without RAG integration (Base Claude). Models were evaluated on withholding AC, withholding and reinitiation timeframes, and platelet and International Normalized Ratio (INR) thresholds. A human assessor scored outputs for guideline concordance (2 = complete, 1 = partial, 0 = none), which were subsequently averaged. RESULTS: Claude achieved the highest overall guideline concordance (1.51/2). Claude, GPT-4o, and Mistral performed similarly in withholding decisions (1.88 [95% CI 1.83, 1.93], 1.86 [1.81, 1.91], and 1.88 [1.84, 1.93]). Claude and GPT-4o showed comparable accuracy in both withholding (1.56 [1.44, 1.67] vs. 1.55 [1.44, 1.67]; p=0.871) and reinitiation timeframes (1.34 [1.20, 1.47] vs. 1.37 [1.26, 1.49]; p=0.920). Mistral achieved the highest platelet threshold concordance (1.52 [1.45, 1.59]), while Claude led in International Normalized Ratio (INR) recommendations (1.37 [1.32, 1.42]). LLaMA (0.76) and Qwen (0.52) significantly underperformed across all categories (p

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Large Language ModelsAnticoagulantsRadiology, InterventionalHumansPatient-Centered CarePractice Guidelines as Topic

Stichwörter der Autor:innen

AnticoagulationArtificial Intelligence (AI)Clinical Decision SupportLarge Language Models (LLMs)Retrieval Augmented Generation (RAG)

Substanzen

Anticoagulants

Autor:innen und Institutionen

  • The Warren Alpert Medical School of Brown University, Providence, RI, USA. hossam_zaki@brown.edu.
  • Massachusetts General Hospital, Boston, MA, USA.
  • Touro University Nevada, Henderson, NV, USA.
  • School of Medicine Department of Radiology, University of Wisconsin, Madison, WI, USA.

Zitieren

APA

Zaki, H. A., Brea, A., Parvataneni, K., Saab, H., Bridge, C. P., Tripathi, S., & Daye, D. (2026). Retrieval-Augmented Language Models for Patient-Centered Periprocedural Anticoagulation in Interventional Radiology. CardioVascular and Interventional Radiology. https://doi.org/10.1007/s00270-026-04428-0

Vancouver

Zaki HA, Brea A, Parvataneni K, Saab H, Bridge CP, Tripathi S, et al. Retrieval-Augmented Language Models for Patient-Centered Periprocedural Anticoagulation in Interventional Radiology. CardioVascular and Interventional Radiology. 2026. doi:10.1007/s00270-026-04428-0

BibTeX
@article{Zaki2026,
  author = {Zaki, Hossam A and Brea, Allison and Parvataneni, Krishnaveni and Saab, Hadeel and Bridge, Christopher P and Tripathi, Satvik and Daye, Dania},
  title = {Retrieval-Augmented Language Models for Patient-Centered Periprocedural Anticoagulation in Interventional Radiology},
  journal = {CardioVascular and Interventional Radiology},
  year = {2026},
  doi = {10.1007/s00270-026-04428-0}
}
Kostenlos starten →