Vetology Rebuilds All AI Classifiers on New Architecture in Two Months

SAN DIEGO, Sept. 18, 2026 /PRNewswire/ -- Vetology, which provides AI radiology screening and on-demand teleradiology consultations to veterinary practices, has rebuilt and re-released all 94 classifiers on a new model architecture.

The team set the rebuild in motion on its own initiative, updating the live library in stages between June and August.

"We used this opportunity to make a radical shift in our model architecture. Our latest models are a significant improvement in that they can interpret global information in a case, both within any single image and between linked images in a study. This follows how a human radiologist reads cases and the success of these models flows directly from that design. They've also opened a wealth of new features that we will continue to build and improve. It was demanding work, and the team was exceptional throughout. Diplomates of the American College of Veterinary Radiology, Drs. John Mattoon and Jennifer Gambino offered contextual guidance and direction as we worked toward improved outcomes for each classifier."

Anthony Lutz, Manager of Data Science

The published performance data describes the classifiers in service now. Every figure belongs to a model running on live cases today.

In March 2025, a joint ACVR and ECVDI position statement in JAVMA named transparency, error reporting, postimplementation monitoring, and unbiased third-party evaluation as conditions for veterinary AI worth trusting. A pilot study in the same journal a year later evaluated six commercial veterinary radiology AI services on general practice canine abdominal radiographs and reported low to moderate performance.

Vetology published condition-level performance data for every classifier in January 2026, and invited independent researchers to test it. That invitation stands.

"We read both of those papers and deliberately addressed all 94 classifiers. Our data science team set a higher bar for this rebuild than the original classifiers were held to, and each one was released only when it met that bar. We believe in being fully transparent about how we develop our AI. That means actively following the published recommendations of accredited institutions and professional bodies, and it means building continuous feedback loops directly into our platform so we can hear from the DVMs using the product every day. When our community raises a concern, we act on it. Our team of board-certified radiologists validates those improvements before they reach our users. We're building a platform veterinarians can rely on and trust, because patient safety is on the line, and that trust is earned through accountability, continuous improvement, and an unwavering commitment to the standards our profession sets."

Eric Goldman, President

Vetology's AI screening reports are built to support the veterinarian's own diagnosis and treatment plan. Cases that call for more can be routed to the company's board-certified radiologists.

"The question I get most is whether the AI report takes the decision away from the veterinarian. It doesn't. The report is one input, it's like asking a colleague to take a look. The diagnosis, the plan, and the conversation with the owner all still rely heavily on the veterinarian's skill and expertise."

Pierre D'Amours, VP of Sales

About Vetology
Vetology is a veterinary diagnostic imaging support company that provides AI-generated screening reports and traditional teleradiology services by board-certified veterinary radiologists. Built by radiologists, Vetology focuses on improving patient outcomes through accuracy, speed, and reliability in diagnostic imaging. Our platform is designed to integrate seamlessly into existing hospital workflows, helping clinicians make informed decisions quickly. Learn more at vetology.net.

Media Contact

Sommer Aweidah
Chief Marketing Officer
Vetology Innovations
888.416.2124
[email protected]

SOURCE Vetology Innovations, LLC