rtificial intelligence (AI) technology is showing big promise in veterinary medicine by improving diagnostic testing, imaging, client communication and drug research.
A prime example of this is AI chatbots and virtual assistants. They offer preliminary assessments and telemedicine services, providing crucial support to urgent care veterinarians by helping prioritize cases. Additionally, they give pet owners immediate guidance, enabling them to make informed decisions when a vet is not readily available.
Dr. Terry Fossum, board-certified veterinary surgeon and CEO of Dr. Fossum’s Pet Care, notes radiology and pathology are two areas that AI has been successfully implemented as a diagnostic tool.
“AI algorithms are used to analyze X-rays, CT scans, MRIs, and ultrasounds, helping veterinarians detect abnormalities such as tumors, fractures, or internal injuries more quickly and accurately than manual analysis,” she explains. “The development of imaging modalities with embedded AI is increasing and may allow for early detection of cancer, which would be potentially life-saving for many of our patients.”
AI is also used in labs for analyzing blood work and other diagnostic tests, and can assist in examining cytology or histopathology slides, detecting patterns indicative of diseases such as cancer or infections.
“The quicker turnaround time benefits patients, as they receive faster, more accurate diagnosis and treatment,” she says. “Veterinarians appreciate being able to complete the patient examination, receive diagnostic results, and implement treatment within a single veterinary visit.”
Another potential risk is protecting data and keeping medical records confidential. Therefore, it is recommended that veterinarians avoid over-relying on AI tools and should always use their professional judgment.
Dr. Sehaj Grewal, DVM, a veterinarian in Los Angeles known as “The Melrose Vet,” echoes this, saying, “AI should only be used as a tool to aid experienced veterinarians but cannot replace the skill and expertise of a vet and their staff. AI systems are not a one-size-fits-all solution. While they can gather and analyze data quickly, AI may struggle with more complex cases or lesser-known symptoms or diseases not represented in the training data. Other limitations include the high cost to implement these technologies, which may make them less accessible to individual vets or smaller practices.”
Dr. Nell Ostermeier, DVM and veterinary advisor at Figo Pet Insurance, notes there are numerous ethical considerations around who owns the data and security associated with the patient and client (pet owner) information.
“There is also the fact that, for humans, AI in medicine is regulated by the FDA, whereas there is no regulatory body in veterinary medicine yet,” he says.
Keep in mind, in order for AI to produce high-quality results, the algorithms need to be fed reliable data in a standardized format. So, the general lack of standardization of medical records and coding in many veterinary practices may create a massive challenge.
Plus, Dr. Fossum adds that implementation of new technology can be a relatively heavy lift.
“Integration with current systems in your practice is one of the main issues that you will face,” she warns. “In addition, training employees can be very challenging and expensive. You need to get the ‘buy-in’ from your entire team to ensure that everyone is using the technology appropriately. They should understand the risks and benefits associated with using AI.”
– Dr. Sehaj Grewal
“AI isn’t perfect and it can cause trust and reliability issues with veterinary staff,” Dr. Fossum says. “A hybrid approach of not totally relying on AI but instead using it as an additional tool may be the most logical way to implement AI now. Over time, veterinarians will likely gain more comfort with the technology.”
“The future of AI will be prevention,” Dr. Grewal forecasts. “AI’s predictive models will be able to help with early disease detection, as well as personalized therapies and treatment plans. Also, AI used in wearables (dog or cat collars) can analyze data from the device in real time, which makes health monitoring more accurate.”
In the future, Dr. Hunt believes AI may be used to assist with medication recommendations for patients with conditions that have multiple medication options.
“Cancer medications could be recommended on the basis of the patient’s demographics, clinical signs, and diagnostic testing,” she says. “These treatments could dramatically improve the effectiveness of existing drug therapies once implemented.”
The potential for AI to transform veterinary practices is significant, but ongoing research and development are crucial to fully understand and harness its capabilities.
“The best way to stay up to date is to read industry trade publications, attend conferences, and sign up for classes on the latest technology,” Dr. Grewal notes. “A good vet is always a student, forever learning and improving on their skills.”