New! Relate by Neo4j | An embedded database for AI agents Learn More

NODES 26 — November 12, 2026

FHIR to Knowledge Graph: Building AI-Ready Clinical Data Models

Session track: Modern Applications

Session time:

Session description:

Healthcare data is fundamentally relational. A patient connects to encounters, encounters to providers, observations to conditions, and conditions to treatments. Yet most clinical software still represents this rich web of relationships as flat records, time-ordered chat logs, or rigid relational tables, forcing both clinicians and AI systems to reconstruct meaningful context manually. In this session, Mansi Parkhi will explore how Fast Healthcare Interoperability Resources (FHIR) data naturally maps to a graph model and walk through the patterns that make this representation powerful. Drawing on production experience building real-time clinical communication platforms, she will show how FHIR resources translate into a property graph, share example data models for patients, encounters, and care journeys, and discuss sample Cypher queries that unlock new capabilities for clinical decision support, follow-up management, and AI-powered triage. You will see data modeling examples, learn about encounter-based history modeling, and walk through lessons learned from designing telemedicine systems where context and relationships matter. Whether you build healthcare software, work with structured clinical data, or design AI systems for regulated domains, this session will give you a practical, code-grounded lens for thinking about clinical data as relationships, not records.

Speaker

photo of Mansi Parkhi

Mansi Parkhi

Software Development Engineer | Amazon | Data & Software Engineering | Applied AI in Business

I am a Software Development Engineer focused on Data and Software. I currently worked at Amazon and also pursuing MBA focused on Applied AI in Business. In addition to my industry role, I am an active researcher and also serve as a reviewer and judge for international conferences and journals. My talks typically focus on practical, experience-driven insights such as operating data platforms at scale and designing robust architectures that move beyond experimentation into production.