EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Comparative Evaluation of Advanced Chunking for Retrieval-Augmented Generation in Large Language Models for Clinical Decision Support

Cesar Abraham Gomez-Cabello iD, Srinivasagam Prabha iD, Syed Ali Haider iD, Ariana Genovese iD, Bernardo G Collaco iD, Nadia G Wood iD, Sanjay Bagaria, Antonio Jorge Forte iD

Bioengineering · 01.11.2025 · 12 (11) · 1194

DOI: 10.3390/bioengineering12111194 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 01.11.2025. Verfasst von Cesar Abraham Gomez-Cabello und 7 weiteren Autor:innen. Beteiligte Einrichtung: Division of Plastic Surgery, Mayo Clinic, Jacksonville, FL 32224, USA. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 76 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Schmidt Family Foundation, Dalio Philantropies, Gerstner Philanthropies.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Retrieval-augmented generation (RAG) quality depends on how source documents are segmented before indexing; fixed-length chunks can split concepts or add noise, reducing precision. We evaluated whether proposition, semantic, and adaptive chunking improve accuracy and relevance for safer clinical decision support. Using a curated domain knowledge base with Gemini 1.0 Pro, we built four otherwise identical RAG pipelines that differed only in the chunking strategy: adaptive length, proposition, semantic, and a fixed token-dependent baseline. Thirty common postoperative rhinoplasty questions were submitted to each pipeline. Outcomes included medical accuracy and clinical relevance (3-point Likert scale) and retrieval precision, recall, and F1; group differences were tested with ANOVA and Tukey post hoc analyses. Adaptive chunking achieved the highest accuracy—87% (Likert 2.37 ± 0.72) versus baseline 50% (1.63 ± 0.72; p = 0.001)—and the highest relevance (93%, 2.90 ± 0.40). Retrieval metrics were strongest with adaptive (precision 0.50, recall 0.88, F1 0.64) versus baseline (0.17, 0.40, 0.24). Proposition and semantic strategies improved all metrics relative to baseline, though less than adaptive. Aligning chunks to logical topic boundaries yielded more accurate, relevant answers without modifying the language model, offering a model-agnostic, data-source-neutral lever to enhance the safety and utility of LLM-based clinical decision support.

Schlagwörter

Stichwörter der Autor:innen

adaptive chunkingadvanced chunkingclinical decision support systemslarge language modelspatient safetyretrieval-augmented generation

Autor:innen und Institutionen

  • Division of Plastic Surgery, Mayo Clinic, Jacksonville, FL 32224, USA.
  • Department of Radiology AI IT, Mayo Clinic, 200 First St. SW, Rochester, MN 55905, USA.
  • Division of Surgical Oncology, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224, USA.
  • Department of AI and Informatics, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224, USA.
  • Center for Digital Health, Mayo Clinic, 200 First St. SW, Rochester, MN 55905, USA.

ORCID-Profile: Cesar Abraham Gomez-Cabello, Srinivasagam Prabha, Syed Ali Haider, Ariana Genovese, Bernardo G Collaco, Nadia G Wood, Antonio Jorge Forte

Förderung

  • Schmidt Family Foundation
  • Dalio Philantropies
  • Gerstner Philanthropies
  • Mayo Clinic
  • Dalio Philanthropies

Zitieren

APA

Gomez-Cabello, C. A., Prabha, S., Haider, S. A., Genovese, A., Collaco, B. G., Wood, N. G., Bagaria, S., & Forte, A. J. (2025). Comparative Evaluation of Advanced Chunking for Retrieval-Augmented Generation in Large Language Models for Clinical Decision Support. Bioengineering, 12(11), 1194. https://doi.org/10.3390/bioengineering12111194

Vancouver

Gomez-Cabello CA, Prabha S, Haider SA, Genovese A, Collaco BG, Wood NG, et al. Comparative Evaluation of Advanced Chunking for Retrieval-Augmented Generation in Large Language Models for Clinical Decision Support. Bioengineering. 2025;12(11):1194. doi:10.3390/bioengineering12111194

BibTeX
@article{GomezCabello2025,
  author = {Gomez-Cabello, Cesar Abraham and Prabha, Srinivasagam and Haider, Syed Ali and Genovese, Ariana and Collaco, Bernardo G and Wood, Nadia G and Bagaria, Sanjay and Forte, Antonio Jorge},
  title = {Comparative Evaluation of Advanced Chunking for Retrieval-Augmented Generation in Large Language Models for Clinical Decision Support},
  journal = {Bioengineering},
  year = {2025},
  volume = {12},
  number = {11},
  pages = {1194},
  doi = {10.3390/bioengineering12111194}
}
Kostenlos starten →