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

Gomez-Cabello, Cesar A.; Prabha, Srinivasagam; Haider, Syed Ali; Genovese, Ariana; Collaco, Bernardo G.; Wood, Nadia G.; Bagaria, Sanjay; Forte, Antonio J.

doi:10.3390/bioengineering12111194

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

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

by

Cesar A. Gomez-Cabello

^1,†

,

Srinivasagam Prabha

^1,†

,

Syed Ali Haider

¹

,

Ariana Genovese

¹

,

Bernardo G. Collaco

¹,

Nadia G. Wood

²

,

Sanjay Bagaria

³ and

Antonio J. Forte

^1,4,5,*

¹

Division of Plastic Surgery, Mayo Clinic, Jacksonville, FL 32224, USA

²

Department of Radiology AI IT, Mayo Clinic, 200 First St. SW, Rochester, MN 55905, USA

³

Division of Surgical Oncology, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224, USA

⁴

Department of AI and Informatics, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224, USA

⁵

Center for Digital Health, Mayo Clinic, 200 First St. SW, Rochester, MN 55905, USA

^*

Author to whom correspondence should be addressed.

^†

These authors contributed equally to this work.

Bioengineering 2025, 12(11), 1194; https://doi.org/10.3390/bioengineering12111194 (registering DOI)

Submission received: 29 September 2025 / Revised: 30 October 2025 / Accepted: 31 October 2025 / Published: 1 November 2025

(This article belongs to the Special Issue Artificial Intelligence in Biomedical Engineering Applications Towards Clinical Translation)

Download Versions Notes

Abstract

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.

Keywords: advanced chunking; adaptive chunking; retrieval-augmented generation; large language models; clinical decision support systems; patient safety

Share and Cite

MDPI and ACS Style

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

AMA Style

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

Chicago/Turabian Style

Gomez-Cabello, Cesar A., Srinivasagam Prabha, Syed Ali Haider, Ariana Genovese, Bernardo G. Collaco, Nadia G. Wood, Sanjay Bagaria, and Antonio J. Forte. 2025. "Comparative Evaluation of Advanced Chunking for Retrieval-Augmented Generation in Large Language Models for Clinical Decision Support" Bioengineering 12, no. 11: 1194. https://doi.org/10.3390/bioengineering12111194

APA Style

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

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

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

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI