How Virginia Tech became the quiet force behind veterinary data

When avian influenza swept through U.S. dairy herds last year, veterinary laboratories across the country were suddenly testing milk samples for a virus they had never analyzed in that way before. Results needed to move fast, not in days, but in hours. New sample types, new testing methods, and new species combinations required entirely new terminology almost overnight.

Behind the scenes, a small team at the Virginia-Maryland College of Veterinary Medicine in Blacksburg stepped in to make that possible.

For more than two decades, Virginia Tech’s veterinary terminology and standards group has been building the infrastructure that allows animal health data to move quickly and accurately between laboratories, federal agencies, and researchers. Most veterinarians have never heard of the team. Yet nearly every major animal health organization relies on the systems they maintain.

Building a common language

One of the biggest challenges in veterinary data isn’t technology — it’s language.

The same disease can be described in dozens of different ways depending on the lab, region, or software system. Without standardized terms, data can’t easily be shared, compared, or analyzed. What one system calls one condition may be labeled entirely differently in another.

Virginia Tech’s group helps solve that problem by maintaining the veterinary extension of SNOMED CT, the global clinical terminology system widely used in human medicine. In hospitals, SNOMED CT ensures that when a physician records a diagnosis, it carries the same meaning no matter where the data is viewed or analyzed. The veterinary extension applies that same concept to animal health.

The work started in the early 2000s when the U.S. Department of Agriculture established the National Animal Health Laboratory Network. As electronic reporting replaced paper forms and emailed PDFs, it became clear that faster communication required more than improved internet connections. It required a shared, structured vocabulary.

At the time, adding new veterinary terms to the global SNOMED system could take more than a year. That timeline didn’t work for emerging disease outbreaks. The Virginia Tech team implemented a faster extension process that allowed new codes to be created in days when new tests, pathogens, or specimen types appeared.

That speed proved essential during the recent avian influenza response, when testing requirements were evolving weekly — sometimes daily.

The system most people never see

The veterinary extension of SNOMED CT — often called VetSCT — doesn’t mean much to most clinicians, and that’s part of the challenge. The group’s work operates quietly in the background, enabling systems to communicate without drawing attention to itself.

But the impact is enormous.

Today, when a National Animal Health Laboratory Network facility identifies a pathogen, electronic messages can transmit standardized data to the USDA within hours. Codes identify the species, sample type, test performed, and result in a way that databases can interpret instantly. That allows disease patterns to be tracked in near real time.

Two decades ago, the process looked very different. Reporting often involved sending documents to Washington, where data had to be manually entered into spreadsheets. It could take months before epidemiologists had a complete picture.

The shift to standardized, electronic reporting has transformed surveillance and response capabilities across the country.

A small team with specialized skills

The group responsible for maintaining this global veterinary terminology is remarkably small. Their expertise sits at the intersection of veterinary medicine, computer programming, and informatics — a rare combination.

That skill set is becoming more valuable as veterinary data expands into new uses.

In 2014, SNOMED’s international governing body recognized Virginia Tech’s veterinary extension as the authoritative global source for veterinary terminology. Requests for new veterinary terms from around the world are now routed through the Blacksburg team.

Why it matters more than ever

The impact of standardized data is expanding beyond disease surveillance.

Several veterinary schools — including Minnesota, Tufts, UC Davis, and Colorado State — are working to transform clinical records into research-ready datasets. By using shared terminology, institutions can compare and combine data across hospitals, creating opportunities for large-scale studies that no single university could conduct alone.

But there’s a challenge: most veterinary records were never structured with this kind of research in mind. Notes written in free text have to be mapped to standardized terms later, which is time-consuming and complex.

That’s why early standardization matters. Structured data from the start makes research faster, more accurate, and more useful.

It also plays a critical role in the future of artificial intelligence in veterinary medicine. AI systems rely on clean, consistent data. When dozens of different names exist for the same disease, machine learning models struggle to produce reliable insights.

Standard terminology helps ensure that data going into AI tools is meaningful and comparable.

A global effort with a local origin

Interest in the veterinary extension has grown internationally. Countries including Australia and South Korea have pushed for easier access, and efforts are underway to make the terminology more widely available through SNOMED’s managed services platform.

While much of the team’s current work focuses on production animal health and food supply protection, there is a broader vision for companion animal medicine.

Only a small number of practice management systems currently support standardized clinical terminology. Most veterinarians still document diagnoses and notes in free text, which makes that information difficult to use for research or data analysis.

Expanding access to standardized coding in everyday clinical practice could open the door to larger shared datasets, better research, and stronger AI tools — but it will require buy-in from software vendors and practitioners alike.

For now, the work continues quietly in Blacksburg. Most veterinarians will never see the terminology systems that support surveillance, research, and data sharing. But every time results move quickly between a lab and a federal agency, or a dataset becomes usable for large-scale study, the influence of Virginia Tech’s team is already there — making veterinary data speak the same language.