Recent Advances in Knowledge Representation and Management in Healthcare: the Emerging Role of Large Language Models

Summary

Objectives: To select and summarize the best papers published in 2024 in the field of Knowledge Representation and Management (KRM) for inclusion in the International Medical Informatics Association (IMIA) Yearbook.

Methods: We conducted PubMed queries and followed the IMIA Yearbook guidelines for performing biomedical informatics literature review to select the best papers in KRM published in 2024.

Results: We retrieved 2,071 publications from PubMed, from which 14 were selected as candidate best papers. After in-depth evaluation, three were selected as the best papers for the KRM section. The candidate papers covered four main thematic areas: (i) Ontology Development and Application, (ii) Knowledge Graph Construction and Completion, (iii) Large Language Model (LLM)-Augmented Knowledge Representation and Reasoning, and (iv) Semantic Interoperability and Data Harmonization. Two of the selected best papers address advances in knowledge graph construction and completion, while the third focuses on semantic interoperability and harmonization across health data standards.

Conclusions: The selection process for the best papers in the KRM section for 2024 showed a wide spectrum of topics. Across several papers, it can be observed the emerging role of LLMs in enhancing ontology engineering, medical knowledge extraction, and reasoning, especially when integrated with structured clinical data and formal knowledge representations.