Artificial Intelligence in Veterinary Medicine

Abstract

 

Up until the past few years, artificial intelligence (AI) was seen as something straight out of a sci-fi movie. AI, however, has rapidly become a reality that will likely be integrated into everyday life. Although the associated benefits and consequences are still relatively unknown, many industries are already experimenting with how AI can help them to gain an edge over competitors by increasing revenue and efficiency—and that includes the veterinary industry.

 

Introduction

 

“AI” refers to the ability of computer systems to perform tasks that normally require human intelligence, and these systems are often hugely complex and powerful with the ability to process unfathomable depths of information quickly.1 Many AI applications are available to the public in a variety of forms, including chatbots such as ChatGPT, art generators such as NightCafe, and AI detectors such as GPTZero—and yet, this is still just a small sampling of what AI technology can provide. AI is therefore likely to significantly change the way in which people and businesses operate in the near future, including how veterinarians provide care to their patients.

 

Industries Exploring Artificial Intelligence

 

AI is being utilized across many service industries as a tool to increase efficiency and revenue. A report by Goldman Sachs, for example, predicts that 44 percent of legal work has the potential to be automated by using emerging AI tools.4 Many university studies have found that AI could affect commodities and investments.4 This, however, presents a substantial conundrum to companies providing legal and investment services. By using AI, each of the previously mentioned industries has the potential to benefit from increased operational efficiency, decreased labor costs, and increased numbers of new clients—which could reduce the number of billable hours that bring in revenue.

 

Insurance companies are also taking advantage of AI. Lemonade Insurance, as an example, entered the industry with AI usage being one of its biggest differentiators and competitive advantages. The first of Lemonade’s three AI programs, a chatbox named AI Maya, helps customers to buy insurance, file claims, and get answers to their customer service questions. Next, CX.AI addresses more complex questions and allows customers to make changes to their insurance policies, increasing Lemonade’s efficiency while empowering customers. The most impressive is AI Jim, which handles the company’s claims. AI Jim can settle a claim in under three seconds and can detect fraud. In fact, AI Jim was able to accurately conclude that one customer was purposefully creating fake accounts and reporting false claims to commit insurance fraud. AI Jim was even able to recognize that this individual was using wigs, hats, and other obstructions to try to block his identity.

 

Artificial Intelligence in the Veterinary Industry

 

AI can also be used for many of the daily tasks commonly performed in veterinary hospitals. For example, receptionists could have more time to answer calls concerning patient needs by having an AI chatbot handle the scheduling of appointments and sending reminders. This also enables receptionists to be more attentive to the clients physically present at the clinic.

 

Perhaps, though, veterinarians have the most to gain from AI usage in the practice. Veterinarians are always strained for time, juggling multiple clients, analyzing lab results and reports, performing surgeries, managing staff, and much more. If an issue arises, it often creates a domino effect for the rest of their day. AI technology, though, can increase efficiency for many tasks and save valuable time for veterinarians due to its massive databases that are continually fed new information to interpret. The AI system then analyzes that data to come to conclusions or make predictions, which can reinforce a veterinarian’s diagnoses and provide better patient care.

 

Dr. Krystle Reagan of UC Davis has been experimenting with AI to generate data analysis of various conditions. Addison’s disease, an endocrinology issue, was her first experiment. As Dr. Reagan describes Addison’s disease, “We call Addison’s disease ‘the great pretender’ because dogs come in with very vague clinical signs. The blood work can look like intestinal disease, it can look like kidney disease, it can look like liver disease. So, it’s one of those conditions that you really have to be on your toes.” This disease often frustrates diagnosticians or results in the wrong diagnosis entirely. However, Dr. Reagan created an AI algorithm by collecting blood samples from over 1,000 patients with AI then able to predict if a patient had Addison’s disease with an accuracy of greater than 99 percent.2

 

Dr. Reagan is now using bloodwork from the past ten years to code data from canine patients to better diagnose Leptospirosis. The current standard is to run two antibody tests ten days apart; 2 waiting those ten days, however, creates a significant delay in starting the treatment (typically dialysis). Being able to find patterns in data sooner could mean that veterinarians would have more certainty in starting treatment before further harm occurs to the patient.

 

Currently, in the veterinary industry, there is no national coded database like there is in human medicine. If AI technology had access to a national or even worldwide database, the system could make more accurate predictions for a multitude of diseases in relation to the specific symptoms of animals. This would create new tools and resources for veterinarians that could increase the potential of preventive medication and the quality of patient care.

 

With an improved database, AI could help in developing more targeted medicines and is a tool that veterinarians, including those with oncology specialties, could use to create individually tailored treatment plans. It is known that patients do not all respond to therapeutics in the same ways, which is why oncologists often use the wait and see approach to determine if their patient is responding to the medicine. Not only is this time-consuming, but it can be costly and stressful. Already, there has been experimentation in which AI has resulted in quicker clinical remission rates:

 

“One application of AI for cancer precision medicine involves the analysis of various drug responses using ‘live’ tumor cells from canine lymphoma patients. This approach, in which researchers use fine-needle aspirates of cancer cells from the affected lymph nodes, uses AI to combine molecular, cellular, and clinical information in order to predict which anti-cancer drugs will work best for a specific dog’s particular lymphoma or leukemia. Researchers tested and analyzed the live tumor cells’ responses to commonly-prescribed chemotherapy drugs using various AI models, and predicted the drugs most likely to work on the patient. Once the prediction report is made to a veterinarian, he or she can design a course of individualized treatment for each patient.”3

 

Artificial Technology Currently in the Veterinary Industry

 

Several products already exist that utilize AI to help veterinarians. Here are a few companies and the services they provide to help veterinary clinics in their everyday operations.

 

Vetology Innovations and SignalPET

 

Reviewing radiographs is a common use of AI in the veterinary industry with Vetology Innovations and SignalPET being two of the biggest companies in the market. The companies use a database with numerous radiographs on file from previous cases. A veterinarian can therefore take a radiograph and immediately upload it for the AI to review. The AI searches and detects any abnormalities and sends the results to the submitting veterinarian in less than five minutes. The AI does not diagnose the patient; instead, it lists abnormalities found in the radiographs and produces an assessment report describing results.

 

Results provided with such a quick turnaround time gives veterinarians the ability to diagnose a patient much more quickly—and a quicker diagnosis could be the difference between life and death for many patients. Many companies also have an option for veterinarians to send the results to their own board-certified radiologist for another opinion.

Many veterinarians who use Vetology or SignalPET have stated that their confidence has increased because of AI supporting their diagnosis. Additionally, having a tool that could potentially save a patient’s life by saving time is tough to argue against. Roland Tripp of the Veterinary Future Society had this to say about AI radiographs. “There is no doubt that the future of veterinary service will incorporate increasing artificial intelligence. X-ray image identification is one of the first ways to market that AI can supplement veterinary practices. In the future, those that welcome AI as just another veterinary tool will generally prosper, and those who wait too long risk being left behind.”7

 

RenalTech

 

RenalTech by Antech Diagnostics utilizes AI for data analysis to predict chronic kidney disease (CKD) in cats, creating a database with more than twenty years of data from more than 730,000 feline patients. It analyzes bloodwork, taking into consideration creatinine, blood urea nitrogen, urine specific gravity, and age values. RenalTech’s AI also compares a patient’s previous blood work and can determine with 95 percent accuracy if the patient will develop CKD within two years. The results have created additional benefits for the clinic. By determining if a patient will likely develop CKD, veterinary clinics can create a revised treatment plan that will help to increase the quality of life for the patient. The patient will therefore potentially benefit from an increased lifespan, decreased kidney symptoms due to a preventive diet change, and closer attention from the patient’s owner. RenalTech’s data has shown that clinics that utilize RenalTech can potentially increase8:

 

  • Average visits from clients who used RenalTech (up 31 percent)
  • Purchasing of CKD food (up 31.5 percent)
  • Purchasing of CKD medicine (up 40.7 percent)
  • Purchasing of blood pressure medicines (up 5 percent)

 

IDEXX

 

IDEXX has created two sophisticated pieces of technology called the SediVue Dx Urine Sediment Analyzer (SV) and the ProCyte One Hematology Analyzer (PO). SV analyzes a patient’s urine to determine the type of bacteria, cells, casts, and crystals that are present in each sample, comparing all sediment values found in the sample to its own database. SV was designed to decrease microscope time for veterinarians and thereby increase efficiency. IDEXX claims that the SV can save the veterinarians an average time of seventeen minutes.9 With nearly twenty minutes saved, a veterinarian could potentially fit in an entire additional appointment and increase their production.

 

The PO works in a comparable manner to the SV but for blood samples. The PO identifies blood cells using five dimensions to get a more accurate characterization of the cells. It also helps mitigate certain conditions such as platelet clumping.10 The PO also considers patterns of results, which help to point veterinarians in a diagnosis direction to detect issues such as lymphopenia.11

 

Zoetis

 

Zoetis created the VetScan Imagyst (VI), which analyzes several sample types such as fecal samples, blood smears, dermatology samples, and cytology ones. The VI then uses its own AI database to detect parasites, yeasts, and inflammatory cells—often doing so within minutes. The VI also generates comparative pictures of the sample and that of its database for the veterinarian to review and compare. Results are also calculated within minutes, and the veterinarian can build a therapeutic plan based off of the results.

 

Limitations and Cautions for Veterinarians Using Artificial Intelligence

 

Though the benefits of using AI products in a veterinary setting may seem like a no-brainer to some, clinics must be careful. Compared to human medicine, the stakes are higher in veterinary medicine because veterinarians can euthanize patients.12 Additionally, the Food and Drug Administration (FDA) does not regulate AI products designed for veterinary use like it does for human medicine.12 This means that there could be substantial liability associated with AI usage due to lawsuits stemming from an overdependence on AI products and associated capabilities.

 

The danger of AI overdependence, of course, can happen across industries. On June 22, 2023 in New York, for example, a federal judge ordered two lawyers and their firm to pay $5,000 for submitting a legal brief that cited multiple fake court cases that were made up by ChatGPT.18 One of the lawyers had utilized the AI to conduct legal research on his behalf and never double-checked any of the fabricated cases. This can be considered a cautionary tale for veterinarians, reminding them to significantly rely on their own education, skills as physicians, and intuition. In other words, AI must not dictate the diagnosis; instead, veterinarians should be using the tools to reinforce and support their own diagnoses.

 

Some veterinarians have been quick to point out the potential problems associated with AI technology in veterinary medicine. As Associate Clinical Professor of Radiology at North Carolina State Veterinary School, Dr. Eli Cohen, stated, “AI image analysis is not the same as a trained radiologist interpreting images in light of an animal’s medical history and unique situation.”12 Currently, AI systems can only identify what they deem to be an abnormality in a radiograph. When a veterinarian consults with a board-certified specialist, they can provide that person with the patient’s medical history so they can put all the pieces together and see the big picture. More importantly, and unlike AI, human consultants have the ability to use their intuition.

 

So, how can veterinarians balance AI usage with human knowledge and experience? As Dr. Cohen puts it, “AI is a powerful tool and will change how medicine is practiced, but the best practice going forward will be using it in line with radiologists to improve access to and quality of patient care, as opposed to using it to replace those consultations.”12 Though AI technology has been designed to streamline human efforts, consultants will still be needed—especially since AI is just being introduced into the industry. Like any other invention, there are typically many undiscovered problems and issues that will arise at the beginning of its use. It is most likely best for veterinarians to be cognizant of the limitations of AI and confirm results with a specialist if they are unsure.

 

Lastly, another possible limitation associated with AI is that it could hinder veterinary advancement. Veterinary advancement comes from new proposed ways of thinking and experimentation, and AI machines are supplied data from the past to predict the future. It will therefore ultimately be up to every veterinarian to continue to innovate and experiment for the good of veterinary medicine. Veterinarians will likely be the ones to drive the industry forward.

Where Do We Go from Here?

With AI being such a relatively new technology, it is difficult to know what directions people will take AI in or what directions AI will take people in. Both the opportunities and risks of AI technology are significant, and perhaps it’s only human nature to focus on the short-term benefits first instead of any potential long-term dangers and consequences associated with AI usage—so it’s important to be aware of these tendencies.

 

As part of the process, people should consider the tech firms that are involved with AI technology and the influential figures within the technology industry that are petitioning to pause AI advancement. Then, thoughtfully review the information received. Most of the articles today about AI are polarized, either describing how beneficial or detrimental AI has the potential to be. There is, however, a balanced way to use AI, so the best option is to proceed with caution and, if possible, to get ahead of any threats that AI could pose.

Final Thoughts

The opportunities and benefits that AI technology could provide have the potential to dramatically and positively change how the world operates at multiple levels. In fact, AI technology is already influencing lives and companies today.

One big question mentioned is whether AI will one day be able to replace doctors, including veterinarians—but, as powerful as AI is, it does have limits. Currently, veterinarians have the competitive edge: they can physically interact with their patients, communicate with pet owners, perform surgeries, and more.

Ultimately, veterinarians should utilize AI technology as a tool. AI presents potential opportunities to provide better patient care, improve efficiency, and provide preventive care. To improve the quality of life of a patient is the main goal of a veterinarian. By utilizing AI, veterinarians will be able to meet and ideally exceed that standard.

 

When used incorrectly, AI could cause great harm to both the patient and the veterinarian. Correctly used AI, though, could be one of the most important tools available for a veterinarian. The clinics and hospitals that embrace and master this technology first will likely have an edge over their competitors—especially if they take a carefully thought out, well-balanced approach.

 

Sources and References

 

  1. Moore, Mike. “What Is AI? Everything You Need to Know About Artificial Intelligence.” TechRadar, 16 Dec. 2019, techradar.com/news/what-is-ai-everything-you-need-to-know.
  2. Nolen, R. Scott. “Artificial Intelligence and Veterinary Medicine.” American Veterinary Medical Association, 15 July 2020, avma.org/javma-news/2020-07-15/artificial-intelligence-veterinary-medicine.
  3. Lim, Sungwon, PhD. “AI — the Newest Tool in Veterinary Science.” IVC Journal, Sept. 2021, ivcjournal.com/ai-newest-tool-veterinary-science.
  4. Mulvaney, E., & Weber, L. (2023, May 11). End of the Billable Hour? Law Firms Get On Board With Artificial Intelligence. WSJ. https://www.wsj.com/articles/end-of-the-billable-hour-law-firms-get-on-board-with-artificial-intelligence-17ebd3f8?page=1.%20Accessed%2012%20June%202023.
  5. Eliot, Lance. “Lawyers Getting Tripped up by Generative AI Such as ChatGPT but Who Really Is to Blame, Asks AI Ethics and AI Law.” Forbes, 29 May 2023, forbes.com/sites/lanceeliot/2023/05/29/lawyers-getting-tripped-up-by-generative-ai-such-as-chatgpt-but-who-really-is-to-blame-asks-ai-ethics-and-ai-law/?sh=427ec5353212.
  6. Hsieh, Nelson. “Lemonade’s AI Jim and Insurance Fraud Detection.” voltequity.com, 30 Mar. 2021, www.voltequity.com/post/lemonades-ai-jim-and-insurance-fraud-detection.
  7. “Products | Veterinary Artificial Intelligence – Vetology.” Vetology, 5 July 2022, vetology.ai/products.
  8. Antech Diagnostics. “RenalTech® for Kidney Disease – Antech Diagnostics.” Antech Diagnostics, 11 Aug. 2022, antechdiagnostics.com/renaltech.
  9. SediVue Dx Veterinary Urine Sediment Analyzer – IDEXX US. (n.d.). IDEXX US. https://www.idexx.com/en/veterinary/analyzers/sedivue-dx-analyzer/
  10. ProCyte One Hematology Analyzer – IDEXX US. (n.d.). IDEXX US. https://www.idexx.com/en/veterinary/analyzers/hematology/procyte-one-analyzer/
  11. Artificial Intelligence in Veterinary Medicine Leads to Efficiency and Superior Accuracy – IDEXX US. (n.d.). IDEXX US. https://www.idexx.com/en/veterinary/analyzers/artificial-intelligence-veterinary-medicine-leads-to-efficiency/
  12. Peake, Tracey. “Artificial Intelligence in Veterinary Medicine Raises Ethical Challenges.” Phys Org, 14 Dec. 2022, phys.org/news/2022-12-artificial-intelligence-veterinary-medicine-ethical.html.
  13. Vallance, B. C. (2023, May 30). Artificial intelligence could lead to extinction, experts warn. BBC News. https://www.bbc.com/news/uk-65746524
  14. CBS Mornings. (2023, March 25). Full interview: “Godfather of artificial intelligence” talks impact and potential of AI [Video]. YouTube. https://www.youtube.com/watch?v=qpoRO378qRY
  15. Center for AI Safety (CAIS). safe.ai.
  16. AI Risk | CAIS. www.safe.ai/ai-risk.
  17. Siddiqui, Sabrina, and Ryan Tracy. “AI’s Rapid Growth Threatens To Flood 2024 Campaigns With Fake Videos.” WSJ, 5 June 2023, wsj.com/articles/ais-rapid-growth-threatens-to-flood-2024-campaigns-with-fake-videos-dbd8144f?page=1.
  18. Weiss, Debra Cassens. “Lawyers who ‘doubled down’ and defended ChatGPT’s fake cases must pay $5k, judge says.” ABA Journal, 26 June 2023, https://www.abajournal.com/web/article/lawyers-who-doubled-down-and-defended-chatgpts-fake-cases-must-pay-5k-judge-says?utm_source=sfmc&utm_medium=email&utm_campaign=&utm_term=&utm_id=688893&sfmc_id=143693123.

 

 

 

 

 

 

 

Skip to content