The Veterinary and Animal Research Data Commons (VARDC) project builds on the success of VetCompass Australia; a collaboration between all seven Australian vet schools, which collates electronic patient records from veterinary practices nationally and aggregates clinical data for researchers to interrogate.

The Veterinary and Animal Research Data Commons project will transform VetCompass Australia to create a platform that is a single point-of-access to multiple, related systems and data types. It will facilitate collaboration with Australia’s leading three veterinary pathology providers to ingest pathology reports and work with the Australian Imaging Service to host images alongside clinical data.

As a world-leading veterinary database, it has advantages over current human public health initiatives in that there are fewer privacy concerns for animals than for humans. This facilitates capacity for development of geospatial disease surveillance and text mining projects that boost human health outcomes.

Start date 1 March 2021
Expected completion date 30 June 2023
Investment by ARDC $393,407
Lead node
1 Sustainable Platform
The creation of a new platform and management portal for selecting, packaging and delivering big electronic patient record data to researchers will include adapting the existing VetCompass Australia system to a new environment and adding a resource that can apply algorithms to the data at scale.
2 The Virtual Lab
This will deliver a framework that shows researchers how to design studies and request data with optimised access and data cleaning. Its central goal will be to generate reproducible epidemiological evidence data that can be comparable across veterinary populations nationally and internationally.
3 Ingestion of pathology reports
This stage will deliver clinical and anatomical pathology data to researchers, linking it to the clinical histories of individual animals in the VetCompass Australia database. This will ensure that data are consistently collected, de-identified and embedded within the correct animal record.
4 Integrations
We will build the capability to establish a connection between the VARDC platform and other data collections. We will also integrate with the Australian Imaging Service to store diagnostic images that can be linked to the clinical records.

Core features

Machine learning and NLP capability
We will use natural language processing to improve and accelerate the searching capabilities of the clinical and pathology data.
Pathology data accessible and linked
The linking of historical clinical records and pathological data from a broad cross-section of the companion animal population will provide a world-first opportunity to understand pathological process across a highly representative animal population, bringing the many benefits of a big-data approach, whilst maintaining a highly detailed view of the individual patients.
Data analysis pipelines
The VARDC platform will house a national “one-stop” virtual laboratory which will offer data analysis pipelines for clinical epidemiological enquiry to researchers.

Who is this project for?

  • Veterinary research organisations
  • Researchers, lecturers and students
  • Professionals working within the veterinary environment, from clinicians to pathology providers
  • Industry professionals (eg vaccine manufacturers)
  • People with companion animals
  • Policy-makers and government

What does this project enable?

The VARDC platform will deliver a framework that shows researchers how to design sample-based and big-data studies and provide optimised access to the data. For the first time, pathology providers and clinical researchers will have a comprehensive data set that includes full clinical histories to assist in setting points of reference and standard intervals, assessing survival data and treatment interventions. By improving the accuracy of their research, disease surveillance will be improved as will health outcomes for pets and improved diagnostic efficiency (economic and social benefit to owners). By capturing search terms and categorising identified pathology data in successive studies we will lay the groundwork for searches enhanced by machine learning.

Handy resources

The University of SydneyVisit
The University of MelbourneVisit
The University of QueenslandVisit
The University of AdelaideVisit
Charles Sturt UniversityVisit
Murdoch UniversityVisit
James Cook UniversityVisit