Veterinary Innovation Council
Ryan B. Appleby, Eleanor Green, David Haworth, and Aaron Massecar
Experts from the Veterinary Innovation Council developed a list of 11 trends that will make the biggest impact on animal health care in the coming years, from artificial intelligence–driven decision support to automatically transcribed patient records. Develop a road map for the future by preparing for the arrival of these innovations and where they may head in years to come.
1. AI for Clinical Decision Support
After reading countless articles on artificial intelligence (AI) in preparation for writing this piece, I shared my findings with a group of veterinary friends while we were out at a local brewery. Most had heard of ChatGPT but few had used it for any veterinary purposes. Out came the phones and the group started tapping away. “Wow. I didn’t expect that,” was a pretty consistent comment as they typed in forms of signalment and asked for treatment plans. Then came, “Yep, that’s what I’d do.” Occasionally the chatbot would get things wrong — sometimes really wrong — but for the most part, the answers mirrored what the veterinarian would do. Then the variations began: “Create a patient handout for postsurgical concerns related to dental extractions.” Bam! What would take at least a few hours to create was complete within seconds. Again, not perfect, and not created on company letterhead, but pretty inclusive of the major concerns. Then the fun really began: patient handouts written in Jane Austen style (signed off, “cordially yours, your obedient servant”) or as a movie script. The possibilities are endless and entertaining. But outside of casual experimentation with friends, there are several ways in which this technology is being leveraged by players in the industry.
2 MONTHS Time it took ChatGPT to gain 100 million monthly active users, making it the fastest-growing consumer application in history. — Reuters
Treatment-Decision Tools: AI decision support tools can analyze a variety of factors, such as the patient’s age, breed, medical history and current condition, to provide personalized treatment recommendations. For example, the Clin.ai platform uses machine learning (ML) to analyze a pet’s medical records and suggest appropriate medications and dosages based on the pet’s individual characteristics.
Challenges and Limitations: While AI decision-support tools have the potential to revolutionize veterinary medicine, there are significant challenges. The first is the need for high-quality data. AI algorithms rely on large amounts of data to make accurate predictions and recommendations, so it is important for veterinarians to ensure that their data is accurate, complete and representative of the patient population. Another challenge is the need for collaboration between veterinarians and AI developers. Developing AI decision-support tools requires a deep understanding of veterinary medicine and clinical decision making, as well as expertise in AI and ML. Finally, it is important to note that AI decision-support tools are not meant to replace veterinarians or their clinical judgment. Rather, they are meant to support the veterinarian in their decision making. — Aaron Massecar, MA, PhD
2. Evolutions in Patient Monitoring
Consumers have become accustomed to the insights that wearable technology can provide in their daily life: “I’m not feeling great, my WHOOP Sleep Score was terrible last night.” Or, “I need to stand for a bit or I’m never going to close the rings on my Apple Watch!” It’s only natural that those interested in gaining better insights into their pet’s health will seek out similar information.
The most advanced at-home patient monitoring systems are wearable devices that allow pet owners to track their pet’s health and activity levels in real time. These devices come in a range of forms, from collars and harnesses to smart beds and cameras, and use a variety of sensors and technologies to monitor everything from heart rate and respiration to activity levels and sleep patterns.
One advanced at-home patient monitoring system is the PetPace Smart Collar, which uses advanced sensors to monitor a pet’s vital signs and activity levels and sends real-time alerts to pet owners and veterinarians if any abnormalities are detected. The PetPace collar’s sensors collect data on the pet’s temperature, activity, pulse, respiration, positions, heart-rate variations and calories consumed and burned.
In addition to wearable devices, there are also advanced at-home monitoring systems that use cameras and sensors to track a pet’s behavior and activity levels. One example is the Furbo Dog Camera, which includes a high-definition camera, two-way audio and a treat dispenser, allowing pet owners to monitor their dog’s behavior and send commands or treats remotely.
The future of at-home veterinary patient monitoring is likely to involve even more advanced sensors and technologies, such as implantable devices that can track a pet’s vital signs and glucose levels in real-time, and AI systems that can analyze this data to detect patterns and identify potential health issues before they become serious.
Additionally, telemedicine platforms and remote monitoring tools are likely to become increasingly important for veterinary care, especially in rural areas or for pets with chronic conditions that require ongoing monitoring and support.
Overall, the most advanced at-home veterinary patient monitoring systems currently available use a combination of wearable devices, sensors and cameras to track a pet’s vital signs, activity levels and behavior in real time and provide pet owners and the veterinary team with valuable insights into the pet’s health and well-being.
As technology evolves, it is likely that these systems will become even more sophisticated, allowing for more accurate and comprehensive monitoring of pets outside of the veterinary clinic. — AM
3. Next-Gen Wearables and the Data They Report
Have you ever wondered if your cat is as judgmental as their looks might suggest? So have the more than 17 million others who downloaded the smart app MeowTalk, which translates what your cat is saying. Cat and dog translators might be too much of a stretch for you, but the other data-generating devices in the home likely won’t be.
4. At-Home Diagnostic Tools
I was at a healthcare information management conference and I heard more than a few speakers talking about the shifting landscape of veterinary care. According to these speakers, 30% of care is moving out of the hospital and into the home. My first reaction: That seems impossible. My surprise and the fact that I surround myself with veterinarians are not unrelated: It seems like most veterinary professionals spend their time focusing on what’s happening in the clinic and not enough time focusing on what’s happening in the home. At-home screening tools have become increasingly popular in veterinary medicine, allowing owners to monitor their pets’ health and well-being from the comfort of their own homes. While these tools can provide valuable insights into a pet’s health status, they also require clinics to prepare for their use to ensure accurate and reliable results. If client demand continues to increase, it is best for veterinary professionals to understand these products so they can better work with and inform pet owners. Here are a few examples of how these tools work.
Blood Glucose Monitors: The AlphaTrak blood glucose monitoring system from Zoetis, designed for cats and dogs with diabetes, requires a small blood sample from the pet’s ear or paw pad, with the results available in seconds. The device also stores past readings, allowing owners to monitor their pet’s glucose levels over time.
Allergy Testing Kits: The Affordable Pet Labs allergy kit includes a collection swab that is used to collect a sample of the pet’s saliva. The sample is then sent to a laboratory for analysis, and the results are available in a few days. The test can identify potential allergies to over 100 different substances, including common environmental allergens and food allergens.
DNA Testing Kits: The Wisdom Panel canine DNA test includes a cheek swab that is used to collect a sample of the pet’s DNA. The sample is then sent to a laboratory for analysis, and the results are available in a few weeks. The test can identify the pet’s breed ancestry and potential health risks based on genetic markers. — AM
3. Choose reliable test kits: It’s important to identify test kits that are reliable and accurate. Look for those that have been validated through clinical studies and have a high level of accuracy.
4. Develop protocols for collecting samples: Set a protocol for how to properly store and transport kits, and make sure your team is trained on these standards.
5. Create a system for tracking and interpreting results: At-home screening tools generate a large amount of data, so it’s important to have a system in place for tracking and interpreting results. Consider using software or other tools to manage this data and ensure that it is accessible.
6. Communicate results with clients: Once you have collected and interpreted the results of at-home screening tools, it is important to communicate these results in a clear and concise manner. Provide recommendations for next steps, such as scheduling a follow-up appointment or making lifestyle changes to improve their pet’s health.
5. Will You Ever Have to Write a Patient Record Again?
I’ve heard that a veterinarian can spend, on average, about two hours a day writing medical records. Some stack those up until the end of the day and either slap the keyboard for a few hours before heading home or wait until they get home, put their kids in bed and work at the kitchen table or couch until late in the night. A few practices have taken the step to hire scribes to take notes and ensure the medical records are up to date.
Then there’s the tech-savvy route: leveraging speech-to-text software. There is no training required — the veterinarian just needs to dictate into a device and the medical record is up to date. Dragon Medical offers a solution and Talkatoo offers a veterinary-specific library that makes transcribing medical terminology more accurate. These tools work as speech-to-text dictation tools, which allow users to speak into their computer’s microphone or an external mic and the tool types out the text as a keyboard extension.
16.6% Share of weekly working hours human medical physicians spend on administrative tasks. — International Journal of Health Services
But what if there was a way to skip the process of hiring and onboarding a scribe or dictating into a machine? These tools already exist in human health care. Products such as Augmedix, Speke and others are fully developed and could have veterinary applications. The promise of these technologies is that they can run in the background as an ambient listening device and the medical record will automatically be populated with the correct information in the proper data fields. Most people who have tried these platforms acknowledge that they’re not perfect — they often require careful review and edits. But, based on the progression of this technology, we could be approaching a time when the doctor simply cues the technology to start a medical record, along with who the doctor is speaking to, and the software will populate the record.
The next step, a step that companies are already working on, is for the software to do more than record the exchange: It will propose a treatment plan based on the signalment. Software like this leverages journal databases (e.g., the Journal of the American Veterinary Medical Association) in order to look at the most up-to-date literature on a particular disease and recommend a course of action based on the findings. These solutions are not widespread, but as William Gibson said, the future is here, it’s just not evenly distributed.
Imagine how the daily schedule of veterinarians might look if writing medical records was a thing of the past and they were able to leverage the latest scientific research for each patient. — AM
6. Diagnostics: Faster, More Accurate and Cheaper?
Take a sprinkle of advanced AI technology and add a heavy dose of medical data points. That is the recipe for powerful diagnostic tools that can assist clinicians in making informed diagnostic decisions and catch easy-to-miss signalment. This emerging technology is already making an impact in three areas, with room for vast growth.
Radiology
AI is being used in radiology to improve the accuracy, efficiency and speed of diagnosis. AI algorithms can be trained to identify patterns in medical images that are nearly invisible to the human eye, helping radiologists diagnose diseases earlier and more accurately.
There are a number of ways that AI is being used in radiology. One way is to develop algorithms that can automatically detect abnormalities in medical images. These algorithms can help radiologists identify potential problems more quickly, which can lead to earlier diagnosis and treatment. Another way that AI is being used in radiology is to develop algorithms that can help radiologists interpret medical images. For example, AI algorithms have been developed that can help radiologists identify the location and extent of a tumor in an MRI scan.
Ultrasonography
AI is being used with ultrasonography in a number of ways to improve the precision, efficacy and speed of diagnosis. AI algorithms can be trained to identify patterns in ultrasound images that clinicians might not be able to catch. This can help radiologists diagnose diseases as soon as possible and more accurately.
Here are some specific examples of how AI is being used with ultrasonography:
- Automated detection of abnormalities. AI algorithms can be used to automatically detect abnormalities in ultrasound images, such as tumors, cysts and fluid collections. This can help radiologists identify potential problems more quickly, which can lead to earlier diagnosis and treatment.
- Quantification of measurements. AI algorithms can be used to quantify measurements in ultrasound images, such as the size of a tumor or the thickness of a heart valve. This can help radiologists make more accurate diagnoses and to plan more effective treatments.
- Image guidance. AI algorithms can be used to guide ultrasound probes during procedures, such as biopsies and needle aspirations. This can help improve the accuracy and safety of these procedures.
- Image interpretation. AI algorithms can be used to help radiologists interpret ultrasound images. This can help improve the accuracy and efficiency of diagnosis.
Cytology
AI is still a relatively new technology in cytology, but it has the potential to revolutionize the way cytology is used in clinical practice. As AI algorithms continue to develop, they will become even more powerful and accurate. As these tools enhance, they will lead to better outcomes for patients and help to improve the quality of animal health care.
Here are some examples of companies that are developing AI-powered cytology solutions:
- Many companies, like IDEXX and Antech, use AI to identify regions of interest but not for diagnosis.
- Zoetis with Vetscan IMAGYST uses automated AI results for fecal, blood and dermatologic samples but not for cytology.
- Most cytology companies are unwilling to provide a specific report with specific results in the way that, for example, SignalPET does for radiology.
As these technologies continue to develop, they have the potential to revolutionize the way that veterinary clinicians progress through a diagnostic workup. — AM
7. Shifts in the Veterinary Business Model
In 2020, I was sitting at a conference, listening to speakers talk about how they evaluated practices before deciding whether to purchase. The standard form of measurement that they used was a multiple of earnings before interest, taxes, depreciation and amortization (EBITDA) that is calculated on a yearly basis. In 2015, it was rare to hear of a practice being sold for more than 15 times EBITDA, but in 2020, some practices were being sold at 30 times EBITDA. Multiple factors contributed to this. For one, consolidators knew that they could come into a practice and create efficiencies of scale through bulk purchasing across a practice group, reduce overhead and run the practice more proficiently, at least on paper. They could then sell the practice because they had increased its overall value. Their goal wasn’t anything other than selling for a higher multiple than what they bought it for.
What’s Next for Practice Ownership
Where might things go in the future? If the valuations keep coming back down to prepandemic levels, then we might see less competition driven by consolidation groups and more opportunities for individual veterinarians looking to purchase a practice. The challenge here is that many graduates are coming out of their programs today without an interest in ownership. That’s where programs like Suveto’s Harbor.vet Student Hospital Immersion Program can play a role. It helps to train the next generation of practice owners through a mentorship model that allows the veterinarian to gradually assume responsibility for running a practice. There are many more like it. For example, Veterinary Practice Partners offers training programs that help with the transition. What we’re likely to see in the coming months is a slowdown in venture capital and an increase in individual practice ownership. — AM
8. The Evolution of Practice Management Systems
There are few situations that cause more consternation in a veterinary clinic than thinking about switching to a new patient management system or practice information management system (PIMS). In the past, startups such as ezyVet and Vetspire came into the market and pushed the envelope with streamlined designs that were cloud-based and therefore more accessible from more locations. Within the past couple of years, ezyVet was acquired by Idexx and Vetspire was acquired by Pathway/Thrive. That doesn’t mean that innovations will cease with either of these products, but historically products tend to become less likely to roll out new features when they’re not trying to capture more market share and compete with other emerging products.However, this trend does not completely inhibit innovation in this sector and there are still many independent options available. For example, Clini.ai, Veterinarian Electronic Assistant and Instinct are all rolling out features that are pushing the PIMS industry forward. Here are a few ways in which PIMSs might advance in coming years.
Telemedicine is also likely to play a bigger role in the future of PIMSs. With telemedicine, veterinarians can offer remote consultations, monitor patients remotely and provide follow-up care without the need for in-person visits. This can be particularly beneficial for patients in rural areas or those with mobility issues.
Another trend that is likely to continue is the focus on client engagement. As pet owners become more tech-savvy, they expect to be able to communicate with their veterinarian more easily and have access to their pet’s medical records and treatment history. By offering online portals and mobile apps, veterinary practices can provide a more convenient and personalized experience.
The future of patient management systems and PIMSs is bright, with many exciting innovations on the horizon. By leveraging AI and ML, adopting cloud-based systems, embracing telemedicine and focusing on client communication, veterinary practices can improve patient care and stay ahead of the curve in an increasingly competitive industry. — AM
9. Enhancing Workplace Culture
There I was many years ago, an educator and scholar at the University of Missouri’s Middlebush Equine Center — and unofficial sounding board for the veterinary nurses/technicians. They would share everything from their career aspirations to the barriers they encountered daily. They loved their positions caring for patients and teaching. It was a fulfilling job. They weren’t complaining, even as they worked tirelessly with salaries that would not sustain them long term. But they wanted someone to inspire them, support them, understand what they do, help them progress in their career and listen. They wanted veterinarians to treat them as the important veterinary professionals that they were. Sound familiar? Today, the profession faces the challenge of attracting and retaining key members of the veterinary team. Innovative solutions are required to face this age-old issue.It would be easy at this point to make the obvious recommendation for all to use veterinary nurses/technicians maximally at the top level at which they can be trained and to reward them financially for doing so. Yes, that is true and necessary. But as we look to the future of practice, we must also consider workplace culture and its impact on empowering every member of the healthcare team.
As leaders, we need to inspire individuals, teams and even ourselves. Leaders are first and foremost responsible for creating and maintaining the desired culture throughout the organization. Today, that culture must include clear signals about the importance of each person on the team, along with integrity, transparency, inclusion and communication. The culture should foster positive attitudes about change that include using veterinary nurses/technicians at the top of their license and taking risks for the purpose of excellence. The environment must be one in which the veterinary nurses/ technicians feel free to make suggestions for improvement, even inspired to do so. They should not only be encouraged, but also rewarded for doing so. They should be acknowledged openly and praised for their positive contributions, even those that are not adopted immediately or those that fail.
As a result, they will feel recognized, as feelings of trepidation, inadequacy and alienation are overcome by expressed belief in what they do and offer. A tradition of contributing ideas, trying new things and being rewarded accordingly becomes contagious throughout the organization.
It is nothing short of amazing to watch organizations evolve favorably as a result of creating an authentically supportive environment in which their people are valued and leaders display empathy.
It is important to recognize that change does not happen overnight. Additionally, most employees do not feel meaningfully connected to the culture of their organization, so patience is important. Leaders must determine and communicate the desired culture, such as through vision and mission. More importantly, they must live the culture themselves, model desired behaviors and motivate others to do the same. — Eleanor Green, DVM, DACVIM, DABVP
10. Vetting Considerations for AI Technology
AI-based systems are now available to veterinarians around the world. However, these products may find themselves in a regulatory vacuum and clinicians should be aware of not only the potential, but also the limitations of such systems. Where AI for human health is regulated by organizations such as the FDA in the U.S., similar regulation is not present in veterinary medicine. Therefore, commercial products enter the market with limited underlying scientific evidence.
3. Ethical Considerations2: Using AI in veterinary practice opens practices up to liability that has not yet been tested in a legal setting. The responsibility likely lies with the attending veterinarian when AI systems make an error. If the underlying performance of these systems is not well understood, veterinarians may not identify inaccuracies prior to a medical error.
Overall, while AI is likely to reshape veterinary practice in the coming years, care should be taken when considering deploying AI in one’s practice at this stage. Practices, clinical groups and organizing bodies should work together to provide guidance in order to best support innovation in practice in a safe and responsible way. — Ryan B. Appleby, DVM, DACVR
11. Precision Medicine in Veterinary Oncology
Precision medicine has revolutionized human medical practice, particularly in oncology. The term “precision medicine” in no way implies that other types are “imprecise,” just that we now have tools that allow us to know exactly what is happening in a specific animal’s disease, and no longer have to draw conclusions based on clinical presentation or history. A clinician in many cases no longer has to play the odds.
Veterinarians and veterinary nurses/technicians in clinical medicine should familiarize themselves with this technology, as it offers huge upsides for patients and practitioners. Potential cases that could benefit from these tests and approaches are seen in general practice every week, and options now exist to keep those cases (plus the client and the revenue) in the clinic — so use them!
When the human genome was sequenced in 2002 (the canine genome quickly following in 2004), the excitement level was at an all-time high. We thought that there would be a few key findings — mountains in the genomic landscape — that would help us unlock the secrets to disease. However, the landscape proved to be more complex than that. It has long been accepted that most, if not all, cancers are caused by mutations in DNA that cause the cell experiencing the mutation to ignore the normal rules and begin to divide uncontrollably. Early in genomics, we thought there would be a handful of mutations common to all tumors that would explain this behavior. It turns out there are hundreds. And many of them interact in complicated ways. Sometimes mutations that impact different molecular pathways result in the same type of tumor. This can be confusing and complex, but thanks to new, cheaper ways to sequence DNA and massive increases in computing power, these landscapes are beginning to be mapped out.
But there is another, more hopeful side to this evolving understanding of cancer’s root causes. If you understand which mutations are driving the growth of a specific tumor (not most of that tumor type, or 72% of that cancer, but that exact tumor, in that dog, on the exam table), then there might be drugs that counter the effect of that mutation. These are called targeted or precision medicines. They target and counteract the effect of the mutation in cancer cells, usually without harming cells that don’t have the mutation. These are very different than traditional chemotherapies that usually attack and kill all rapidly dividing cells. There are more than 145 of these therapies approved by the FDA for use in humans. There is usually extensive safety data for these drugs in dogs, generated as part of the FDA approval process. Efficacy data in dogs is being generated all the time. So the combination of these three factors — an increasing understanding of the genomics behind cancer, an increasing ability to detect the mutations driving a tumor’s growth, both combined with increased availability of drugs that target the exact mutations in those tumors — is what makes it such an exciting time in oncology. There are veterinary companies that can quickly assess the mutations driving a specific dog’s cancer and, once identified, recommend targeted therapies that are safe and likely efficacious.
The tests and therapies aren’t cheap, but because they could eliminate much of the expense and inconvenience of traditional chemotherapy, the total cost of this approach might actually be less expensive and have higher compliance rates for cancer treatment than what we’ve become used to in the past. We are in early days, but this is a very rapidly evolving field due to the enormous resources being committed in human health care, and inevitably availability will go up as prices go down. — David Haworth, DVM, PhD