Artificial intelligence (AI) is in the news everywhere! But what does artificial intelligence use in veterinary medicine look like today, and what could it look like in the future? To answer those questions, we talked to Parminder Basran, PhD, a Fellow of the Canadian College of Physicists in Medicine. Basran is an Associate Research Professor, Section of Medical Oncology, in the Department of Clinical Sciences at Cornell University’s College of Veterinary Medicine.
Basran was quoted by Cornell as saying: “These articles provide a great starting point for anyone who wants to learn about AI and veterinary medicine. While much of the focus is on diagnostic imaging, many of the articles cover topics broad enough to provide a general understanding of the AI-related subjects. They’re an accessible and readable set of articles that the community will (hopefully) enjoy.” (Editor’s note: See below for links to these articles.)
How Can AI Help Veterinarians
There are a number of ways you can leverage AI to help veterinarians,” said Basran. “I think a big one is image analysis. Another is diagnostic decision support tools. There’s an emerging field of drug discovery. And finally there’s predictive modeling.”
With regard to image analysis, Basran said we can use AI algorithms to help find features within things like X-rays and CT scans. This can help analyze these images.
“To date, there are a couple of commercial entities that provide AI-based medical image analysis largely to help radiologists to get through a lot of their caseloads,” said Basran. “There is a shortage of veterinary radiologists, so this is an option for a lot of communities.”
Another area is diagnostics. Basran said, “We are seeing AI algorithms being developed to diagnose animals with various conditions using datasets that can rage anywhere from age, sex, weight to whole-slide images, biospecimen samples like blood smear images, infrared and conventional imaging, GPS sensors, and all sorts of novel ways to collect data from animals that can be used in decision support in diagnostics.”
For drug discovery on the commercial side, Basran said they can use machine learning techniques—which is a sub-field of AI—to help analyze huge sets of data that people collect. This analysis could lead to being able to discover drugs to treat specific conditions.
Predictive modeling can be used to AI algorithms to predict various conditions. “In the case of equine medicine, there’s been some research on developing models to predict laminitis, catastrophic fractures of proximal sesamoid bones, and other kinds of conditions,” said Basran.
To hear more about artificial intelligence use in veterinary medicine, listen to the complete podcast. It’s pretty amazing stuff!
About Parminder Basran, PhD, FCCPM
Parminder Basran is a PhD as well as a Fellow of the Canadian College of Physicists in Medicine. He is an Associate Research Professor, Section of Medical Oncology, in the Department of Clinical Sciences at Cornell University’s College of Veterinary Medicine. He has more than 20 years of experience as a Medical Physicist in Human Oncology. Basran said he was excited to bring his medical physics knowledge to Cornell through research, education and clinical support. He said while the patients are different, “my ethos remains the same: to improve the quality of care for cancer patients through the introduction of evidence-based technologies and pragmatic processes in our fast-paced environment.”
The open access articles, entitled “Artificial intelligence 101 for veterinary diagnostic imaging,” “Radiomics in veterinary medicine: Overview, methods, and applications,” and “The role of artificial intelligence in veterinary radiation oncology,” are meant to give a broad overview on the topic.
You can read more from the Cornell interview of Basran at https://www.vet.cornell.edu/news/20230106/new-horizons-artificial-intelligence-veterinary-medicine.