Patient-centric knowledge graphs: a survey of current methods, challenges, and applications

Area of Science:

• * Health Informatics
• * Artificial Intelligence in Medicine
• * Knowledge Representation

Background:

• * Healthcare data is often fragmented across disparate systems.
• * A unified patient health perspective is crucial for individualized care.
• * Patient-Centric Knowledge Graphs (PCKGs) emerge as a solution.

Purpose of the Study:

• * To review methodologies, challenges, and opportunities in PCKG development.
• * To explore PCKG applications in personalized medicine and disease prediction.
• * To provide a foundational perspective on the state-of-the-art in PCKGs.

Main Methods:

• * Literature review of PCKG methodologies.
• * Analysis of data integration, knowledge extraction, and representation techniques.
• * Exploration of reasoning, semantic search, and inference mechanisms.

Main Results:

• * PCKGs integrate diverse health data for a comprehensive patient view.
• * Advanced techniques enhance PCKG construction and evaluation for actionable insights.
• * PCKGs show significant potential in personalized medicine and treatment planning.

Conclusions:

• * PCKGs represent a significant shift towards individualized patient care.
• * Effective PCKG development requires addressing complexities in ontology design and data integration.
• * Future research should focus on optimizing PCKG applications for improved healthcare outcomes.