I lead CSIRO’s work on health data analytics using
FHIR and clinical terminology, for research, clinical
decision support, data quality and AI. I created
Pathling and lead the HL7 SQL on
FHIR project. I have worked in Australian digital
health since 2010, including as lead architect of the National
Clinical Terminology Service.
An open-source toolkit for analysing FHIR data with
clinical terminology, built on Apache Spark, with libraries for
Python, R, Java and Scala.
Python package downloads grew from about 3,000 to about 20,000 a
month between mid 2025 and mid 2026.
Used by research and health organisations including the
University of Queensland, SickKids in Toronto, the Murdoch
Children’s Research Institute and the Bavarian Cancer Research
Center.
The query engine behind CSIRO work on
FHIR-to-OMOP transformation, data
quality assessment and agent-based analytics.
An HL7 specification for portable, tabular views of
FHIR data, defined using FHIRPath. A view runs
unchanged on any conforming engine, from DuckDB on a laptop to a
Spark cluster.
Thirteen independent implementations, including those from
Google, Health Samurai, Medplum, NCQA and
SAS, are tested against a shared conformance suite.
Took the specification to its September 2026 HL7 ballot,
including new operations for running and exporting views and
SQL queries.
Most active contributor to the specification, and first author
of the paper that describes it.
Wrote an implementation that compiles views to T-SQL
for Microsoft SQL Server.
Experience
CSIRO Australian e-Health Research Centre
– present
Technical Lead, Health Data Standards and Analytics (Principal
Research Consultant)
Building Signet, an open-source SMART on
FHIR authorisation server with passkey
authentication.
Co-developed the open-source FHIR Bulk Client for
Java, a client for the Bulk Data Access specification.
Led Aus CVD Risk-i, which brought the Heart
Foundation cardiovascular risk calculator into general practice
systems using SMART App Launch and Smart Forms.
Contributed to Aged Care Data Compare, Smart Health Checks, the
National Approach to Genomic Information Management, Enabling
FHIR in Queensland Health and sBeacon, a scalable
genomic data exchange.
The next three roles ran concurrently, contracted through my
company, Small Spark.
Australian Digital Health Agency
–
Lead Architect, Clinical Terminology (NEHTA until
2016)
Lead architect of the National Clinical Terminology Service: the
national portal, FHIR terminology server and
syndication service for SNOMED CT-AU, the
Australian Medicines Terminology and other content.
Built the open-source Polecat and Hotspot terminology browsers,
and contributed to Medserve, a FHIR server for
Australian medicines terminology.
Built Serge, which imported and reconciled TGA and
PBS data feeds to help terminology analysts keep
the medicines terminology current.
Helped establish the Agency’s open source program, gaining
approval to release six existing code bases under Apache 2.0.
Mojo Soup
–
Consultant
Built the back end of Soup Mail, a hosted service for sharing
email within teams, and ran it in production. Ruby on Rails, Solr,
AWS and Puppet.
Central Petroleum and Santos
–
Lead Consultant and Business Analyst
Led the information systems side of the Mereenie oil and gas field
handover to Central Petroleum. Requirements and integration
testing for field management and gas pipeline systems at Santos.
National E-Health Transition Authority
–
Business Analyst, national e-health record
Requirements and stakeholder engagement for the personally
controlled electronic health record, now My Health Record.
Facilitated agreement with the pathology sector on how result
reports would be sent to the record, and helped design the
national service for clinical document templates.
Earlier
Web developer, thrudigital, London
Business analyst, data warehousing, Suncorp–
Business analyst, business intelligence, Queensland
Government–
Information officer and business analyst, Department of
Natural Resources and Water–
Independent work
Small Spark, my own company, building Ruby on Rails
e-commerce sites–
TypeFront, font hosting for the web with licensing controls for type
foundries–
Blavel,
travel blogging with geocoded posts and maps–
Rosettasaurus, a community-edited dictionary of translated sentences
Standards and speaking
I represent CSIRO at HL7 Working Group Meetings,
contributing to the FHIR specification and its tooling,
including the FHIR IG Publisher. I ran the
SQL on FHIR track at the HL7 International
connectathon in Rotterdam and have led tracks at HL7 Australia
connectathons.
Speaker at MEDINFO, HL7 FHIR DevDays,
SNOMED CT Expo and HL7 connectathons.
Publications
Papers
Grimes J, Brush R, Rhyzhikov N, Szul P, Mandel J, Gottlieb D, et
al.
SQL on FHIR - Tabular views of
FHIR data using FHIRPath.
npj Digital Medicine. 2025;8:342.
doi:10.1038/s41746-025-01708-w
Grimes J, Szul P, Metke-Jimenez A, Lawley M, Loi K.
Pathling: analytics on FHIR. Journal of
Biomedical Semantics. 2022;13:23.
doi:10.1186/s13326-022-00277-1
Software
SQL on FHIR views for Microsoft
SQL Server
(2025). Compiles views to T-SQL.
doi:10.25919/vnzg-tn91