Home // AIHealth 2024, The First International Conference on AI-Health // View article

Medical Knowledge Harmonization: A Graph-based, Entity-Selective Approach to Multi-source Diagnoses

Authors:
Andrea Bianchi
Antinisca Di Marco

Keywords: medical diagnostics, multi-source diagnosis

Abstract:
The paper discusses a novel system for medical diagnostics that integrates patient data from various sources to address the fragmentation of healthcare information. By generating and merging knowledge graphs from raw medical texts focused on key biomedical entities (Gene, Disease, Chemical, Species, Mutation, Cell Type), the system facilitates a comprehensive understanding of a patient's medical history. It accurately extracts and connects critical entities, creating individual and combined knowledge graphs that elucidate a patient's medical journey. This approach helps bridge diagnostic gaps, offering a visual tool for practitioners to detect patterns and discrepancies in patient data. Despite limitations such as language dependency and validation scope, this system sets the stage for future enhancements toward a more universally accessible and clinically useful healthcare system.

Pages: 35 to 40

Copyright: Copyright (c) IARIA, 2024

Publication date: March 10, 2024

Published in: conference

ISBN: 978-1-68558-136-7

Location: Athens, Greece

Dates: from March 10, 2024 to March 14, 2024