The next time your cat has x-rays, don’t be surprised if a computer with artificial intelligence reads them. X-rays for cats are going even higher-tech, largely because there’s a shortage of human veterinary radiologists. While AI gets mixed reviews from vets, a growing number of practices are using it. Your vet might be next!
We’re all for transparency, so we want you to know clicking on some links in this post will take you to Amazon. We only recommend products we trust.We make a small commission for purchases made through this blog at no additional cost to you. The commissions help keep this blog running and the cats fed!
Don’t get me wrong. I’m all for technology in veterinary care. But I’m not sure about artificial intelligence replacing radiologists. I think I’ll always prefer having x-rays for cats read by a human, not a machine. Not that we have a choice. The American Veterinary Medical Association says there aren’t enough radiologists or specialists of any kind.
It used to be that universities hired most of the board-certified specialists. They taught, did research and treated animals in the universities’ teaching hospitals. But things are different now. Now, most specialists work in industry or private practice where they can earn more money and focus on patient care.
In 2013, board-certified specialists in private practice earned an average of $180,000 a year, the AVMA says, while doctors at universities averaged $151,000 a year.
The differences in job descriptions and salaries leave the universities short of faculty to teach future specialists, and the need is especially acute in radiology.
Today, radiologist Seth Wallack says in an article for Veterinary Practice News, there are fewer than 35 veterinary radiology programs in the US, and each graduates just 30-40 specialists a year.
“Unfortunately, as we continue to produce more radiographs than specialists can read, turnaround times get longer and longer,” delaying diagnoses, treatments and further testing, he says.
How Artificial Intelligence Reads X-Rays For Cats
In his article for Veterinary Practice News, Wallack suggests thinking of AI as “transferring a radiologist’s experience and knowledge to a computer chip that doesn’t get tired by the end of the day.” When used in a vet’s office, it can speed up diagnostics, although, like human radiologists, it can make mistakes.
One vet told Veterinary Information Network News AI “sometimes identified lesions that were not really there.” And another said “I think it’s helpful, but it has its limitations. Right now, the technology will in no way, shape or form replace a radiologist’s report.”
And computer chips, it turns out, are not exactly quick studies. It took hundreds of thousands of radiographic images to teach the diagnostic tools AI uses to identify the abnormalities that may indicate disease. Reviews by radiology experts and feedback from users improved the tools’ accuracy.
“With ongoing work,” Wallack says, “an AI algorithm can be trained to be as good as a human radiologist.” In his Veterinary Practice News article, he sites US Food and Drug Administration online documentation that puts accuracy for human radiology in the high 80s to low 90s. He assumes it would be the same for animals.
Don't Rule Out The Radiologist, At Least Not Yet
So far, just two companies in the US — SignalPET and Vetology AI — are offering the technology, and they say thousands of practices are buying it. SignalPET, which launched last year, was founded by Neil Shaw, who with his brother Darryl, founded the emergency and referral hospital chain, BluePearl. They sold the hospitals to Mars, Inc. in 2015.
Vetology AI is part of Vetology, founded in 2010 by Wallack as a traditional teleradiology service. But Wallack says the AI business is growing “in leaps and bounds.” It currently serves 2,935 clinics with more coming onboard monthly.
Does that mean human radiologists should consider another line of work? Probably not. “The truth is AI and radiology specialists go hand in hand and work best together,” Wallack says.
Read More Veterinary News