Kitty CSI: How One Lab Is Investigating Cat Deaths and Using AI to Fight Animal Cruelty

Most people have experienced it at least once. You’re driving along and notice a small, still shape on the side of the road. You feel that familiar sinking feeling, then keep moving, trying not to think about it.

At the University of Florida, there’s a team that does the opposite. They stop, investigate, and work to understand exactly what happened.

At UF’s Veterinary Forensic Sciences Laboratory, veterinarians and researchers study the deaths of free-roaming cats to determine causes of death and, in some cases, uncover evidence of animal cruelty. The program receives deceased cat specimens from across the country, building a growing database that helps identify patterns, improve investigations, and support legal cases when needed.

Leading that work is Adam Stern, a professor of forensic pathology and veterinarian whose team performs detailed postmortem exams, analyzes environmental toxins, and reconstructs the circumstances surrounding each death.

“We look at everything,” Stern said. “We’ll always take a history. Someone might say they saw a person put food down, and then the cat died, but we always do our own exams and look for everything. Is there blunt trauma? Is there natural disease? We look at everything, because if you go in with tunnel vision, there’s a big chance you’re going to be wrong.”

The lab estimates time of death, tests for environmental contaminants like lead, and evaluates injuries to determine whether trauma was accidental, natural, or potentially caused by a person. In some cases, Stern’s work brings him into the courtroom to provide expert testimony.

Now, the work is expanding beyond Florida through an international collaboration powered by artificial intelligence.

Jon Kim, D.V.M., Ph.D., joined the University of Florida as an assistant professor in 2022 through its Artificial Intelligence initiative. His research focuses on developing diagnostic and clinical tools using AI and machine learning, particularly in comparative oncology and translational medicine. When he learned about the forensic lab’s growing caseload and data collection, he saw an opportunity.

“I realized that AI could be very helpful here, in its application to forensic science as well,” Kim said. “They are dealing with a lot of data; they are writing reports, analyzing pieces of DNA, incorporating pathology imaging. This would help them advance their scientific work.”

The collaboration quickly extended internationally. South Korea, which has a large population of free-roaming cats, has been working to strengthen its own veterinary forensic capabilities. But with limited specialized training and growing case numbers, officials were looking for ways to make investigations more efficient and consistent.

Stern and Kim established a research partnership with South Korea’s Animal Disease Diagnostic Division of the Animal and Plant Quarantine Agency. Each year, pathologists travel to Gainesville to train and collaborate with the UF team. Stern also visited South Korea in 2025 and plans to return again.

A key part of the collaboration focuses on standardizing how forensic data is collected and analyzed so findings can be compared across regions.

“Different people from different places tend to do science a little bit differently,” Kim said. “Creating a standard scoring system becomes important for validating the data.”

AI is playing a growing role in that effort. By analyzing large datasets — including pathology images, DNA findings, and case histories — machine learning tools may help identify patterns that suggest whether a death was accidental, natural, or suspicious. For example, the technology could assist in distinguishing between injuries caused by vehicles and those caused by other objects.

While AI will never replace the hands-on expertise of a trained forensic pathologist, it could help prioritize cases and flag those that require closer investigation.

“The data that comes from populations with cancer is very large, but so is the data that comes from even one individual,” Kim said. “The data is very complex, and something like AI can make it more manageable to synthesize it and discover patterns.”

Looking ahead, Stern believes that kind of technology could eventually help professionals who don’t have the time or resources to perform full forensic autopsies on every case by identifying those that may warrant deeper review.

At its core, the work is about more than science or technology. It’s about understanding what happened to animals that often lived unnoticed and ensuring that, when necessary, their deaths are taken seriously.

For the researchers behind the project, each case is part of a larger mission: to improve animal welfare, support investigations, and bring clarity to questions that might otherwise go unanswered.