Knowledge Graphs Extracted from Medical Appointment Transcriptions: Results Generating Triples Relying on LLMs

Rafael Roque de Souza, Thiago Pinheiro, Julio Oliveira, Julio Reis

2023

Abstract

Knowledge Graphs (KGs) represent computer-interpretable interactions between real-world entities. This can be valuable for representing medical data semantically. We address the challenge of automatically transforming transcripted medical conversations (clinical dialogues) into RDF triples to structure clinical information. In this article, we design and develop a software tool that simplifies clinical documentation. Our solution explores advanced techniques, such as the Fine-tuned GPT-NeoX 20B model, to extract and summarize crucial information from clinical dialogues. We designed the solution’s architecture, supported by technologies such as Docker and MongoDB, to be durable and scalable. We achieve accurate medical entity detection from Portuguese-language textual data and identify semantic relationships in interactions between doctors and patients. By applying advanced Natural Language Processing techniques and Large Language Models (LLMs), our results improve the accuracy and relevance of RDF triples generated from clinical textual data.

Paper Citation

in Harvard Style

Roque de Souza R., Pinheiro T., Oliveira J. and Reis J. (2023). Knowledge Graphs Extracted from Medical Appointment Transcriptions: Results Generating Triples Relying on LLMs. In Proceedings of the 15th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 2: KEOD; ISBN 978-989-758-671-2, SciTePress, pages 129-139. DOI: 10.5220/0012259000003598

in Bibtex Style

@conference{keod23,
author={Rafael Roque de Souza and Thiago Pinheiro and Julio Oliveira and Julio Reis},
title={Knowledge Graphs Extracted from Medical Appointment Transcriptions: Results Generating Triples Relying on LLMs},
booktitle={Proceedings of the 15th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 2: KEOD},
year={2023},
pages={129-139},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012259000003598},
isbn={978-989-758-671-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 2: KEOD
TI - Knowledge Graphs Extracted from Medical Appointment Transcriptions: Results Generating Triples Relying on LLMs
SN - 978-989-758-671-2
AU - Roque de Souza R.
AU - Pinheiro T.
AU - Oliveira J.
AU - Reis J.
PY - 2023
SP - 129
EP - 139
DO - 10.5220/0012259000003598
PB - SciTePress