Leveraging LLMs and RAG for Schema Alignment: A Case Study in Healthcare

Rishi Saripalle, Roopa Foulger, Satish Dooda

2025

Abstract

In the quest to achieve digital health and enable data-driven healthcare, health organizations often rely on multiple third-party vendor solutions to monitor and collect patient health and related data, specifically outside organizations' control, such as home setting, which is later communicated to the organization’s information systems. However, the reliance on multiple vendor solutions often results in fragmented data structures, as each vendor solutions system follows its non-standard data model. This fragmentation complicates the data integration, creating barriers to seamless data exchange and interoperability, which is essential for data-driven healthcare. Recent advancements in Large Language Models (LLMs) have great potential to analyze data models and generate rich contextual-semantic metadata for the model, useful for identifying mappings between disparate data structures. This preliminary research explores the adoption of LLMs in combination with the Retrieval-Augmented Generation (RAG) approach to facilitate structural alignment between disparate data models. By semi-automating the schema alignment process—currently a labor-intensive task—LLMs can streamline the data integration of heterogeneous data models, enhancing efficiency by reducing the developer’s time and manual effort.

Paper Citation

in Harvard Style

Saripalle R., Foulger R. and Dooda S. (2025). Leveraging LLMs and RAG for Schema Alignment: A Case Study in Healthcare. In Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 2: HEALTHINF; ISBN 978-989-758-731-3, SciTePress, pages 750-757. DOI: 10.5220/0013262800003911

in Bibtex Style

@conference{healthinf25,
author={Rishi Saripalle and Roopa Foulger and Satish Dooda},
title={Leveraging LLMs and RAG for Schema Alignment: A Case Study in Healthcare},
booktitle={Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 2: HEALTHINF},
year={2025},
pages={750-757},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013262800003911},
isbn={978-989-758-731-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 2: HEALTHINF
TI - Leveraging LLMs and RAG for Schema Alignment: A Case Study in Healthcare
SN - 978-989-758-731-3
AU - Saripalle R.
AU - Foulger R.
AU - Dooda S.
PY - 2025
SP - 750
EP - 757
DO - 10.5220/0013262800003911
PB - SciTePress