How Artificial Intelligence is Changing the Veterinary Science & Livestock Sector : A Latest Review
Veterinary practices are busier than ever, which is a good problem to have. But do you ever feel like you’re choosing between practicing high-quality medicine and simply managing your case load? Advances like point-of-care testing and digital radiography can be lifesavers, because they expand your capabilities and improve efficiency. But when making these purchases, it’s important that your choices fit your practice’s needs and help you maintain the high clinical standards your clients and patients deserve.
Artificial intelligence (AI) incorporates a large range of concepts and technologies used to solve problems of logical or algorithmic complexity. Introduced in the 1950s, many AI methods have been developed or extended recently with the improvement of computer performance. Artificial intelligence crosses many categories, including mechanistic modelling, software engineering, data science and digital statistics. Recent developments have been fuelled by the interfaces created between AI and other disciplines, such as bio-medicine, as well as massive data from different fields, particularly those associated with healthcare services . AI was legitimately new to the animal health care sector. Researchers, scientists, and entrepreneurs have begun to introduce machine learning and AI into veterinary health. A key consideration in the successful adoption of AI technologies in veterinary medicine will be their safe introduction into clinical practice. They propose a 4-phase approach to follow when introducing an AI based solution in the hospital includes, conceptualization of the AI project, data acquisition and preparation, AI application and translation . There is an emergent demand for the development and use of wearable’s, smart cameras and sensor devices in animal health for pet animals and on farms . These devices generate an enormous amount of data, which increases the potential for artificial intelligence (AI) using machine learning algorithms and real-time analysis (RTA). Perusal of literature revealed that very limited reports were published with special references to understanding of technologies (AI and machine learning) within the animal health industry.
While veterinarians are busy juggling time management, patient care, and other aspects of practice, the demand for services can compete with the need to offer timely, affordable care. Artificial intelligence (AI) technology is just one example of how these goals can be pursued simultaneously. If you still think of AI as science fiction, you may be surprised to know that it’s already changing human medicine; and now, AI offers many of the same advantages to veterinarians.
Artificial intelligence (AI) is a branch of computer science in which computer systems are designed to perform tasks that mimic human intelligence. Today, AI is reshaping day-to-day life and has numerous emerging medical applications poised to profoundly reshape the practice of veterinary medicine. In this Currents in One Health, we discuss the essential elements of AI for veterinary practitioners with the aim to help them make informed decisions in applying AI technologies into their practices. Veterinarians will play an integral role in ensuring the appropriate uses and good curation of data. The expertise of veterinary professionals will be vital to ensuring good data and, subsequently, AI that meets the needs of the profession.
Artificial intelligence (AI) is fairly new to the animal health sector. Over the past decade, however, researchers, scientists, and entrepreneurs have begun to introduce machine learning and AI into veterinary health, with groundbreaking results.
AI is not always welcomed with open arms when it appears in an industry. People often have concerns. They worry that artificial intelligence will replace humans and remove jobs. However, this is not the case in veterinary science. When it comes to diagnosing and treating diseases in animals, nothing can replace the expertise of a good veterinarian. AI is a valuable, advanced tool that can help veterinarians do their jobs with greater speed and precision, freeing them up to focus on complex tasks and concentrate on the welfare of their patients.
Artificial intelligence (AI) is an important technological advancement changing the shape of our lives. Employed by many of the largest companies in the world, AI technologies power our favorite mobile applications; offer movie, TV, and music suggestions; and predict the next word in our text messages. The opportunities to leverage the power of AI to improve our quality of life in our day-to-day or professional activities seem endless, notwithstanding ethical challenges. Nowhere is this opportunity more apparent than in medical practice. Over the last 2 decades, publications on AI in medicine have increased exponentially.1 In particular, there have been profound advances in AI and the field of diagnostic imaging where technologies have been developed to aid in diagnosis and support radiologists in both research and commercial settings. A similar shift is occurring in veterinary medicine: we are on the cusp of a large and unpredictable shift in available technology that has the potential to reshape how veterinary medicine is practiced.
Practically speaking, it is not necessary for the average veterinary practitioner to have a working knowledge of computer programming to effectively use and implement AI. However, the implementation of adopting a technology can be considered analogous to implementing a new diagnostic test in practice. For example, the average veterinarian cannot design and develop an ELISA for benchtop testing. However, veterinarians do have enough underlying knowledge of what an antigen or antibody is, as well as how ELISA testing works to feel confident employing so-called SNAP tests in practice. Furthermore, veterinarians are aware of the need to have a quality management system in place to ensure the tests work as expected. Similarly, while details of the computer algorithm designed for diagnosis or detection of diseases will initially be foreign to most veterinarians, a baseline level of knowledge is needed to understand the power and pitfalls of AI.
Artificial intelligence is still in its nascent stages but will likely have a profound impact on our profession in the years to come. Therefore, it is vital that all veterinarians understand both the promise and limitations of AI.
Potential use of artificial intelligence in veterinary care In the last few years, AI has brought great reforms to the veterinary industry by making, veterinary diagnostics easier, medical care accessible data collection for the pet industry. a. Pet trackers and pet cameras: Pet trackers and pet cameras monitor virtually all of your pet’s daily activities such as, movement, eating and drinking behavior, sleep patterns, etc. This vast amount of information opens up many possibilities for machine learning [10]. Using video analytics software that is programmed to be on the lookout for unusual behavior, the camera sends you an alert to check for an injury or a foreign object causing discomfort. Similarly, a pet tracker detects that your cat has been sleeping several hours more each day than
normal. An algorithm in the tracker texts you that something may be amiss with your kitty and that you may want to call your vet . b. Pet telehealth: Veterinary telehealth, the use of technology to deliver vet care, can benefit from the use of chatbots. A chatbot is software that simulates a natural human conversation (either written of spoken). Vet clinics can install chatbots on their websites to act as the first level of interaction with customers. In addition to assisting with communication, advanced bots could answer pet health questions and analyze symptoms . A pet version of the “Ada Health app” could be uses a conversational interface to determine symptoms and provide medical information. If needed, the app then offers a remote consultation with a real doctor. c. Dog walking: Dog walking companies like Rover and Wag could benefit from a natural language processing (NLP) system that delivers a weekly recap to their customers about their pet’s activities. Dog owners could receive a personalized email story detailing the walk routes, duration, encounters with other dogs and food and potty breaks . d. Smart feeder: Smart feeder could determine when pet food is running low and suggest a reorder, mentioning current coupons or special discounts. Smart feeders could also monitor a pet’s eating habits and highlight any irregularities .
Understanding Artificial Intelligence
When machines are able to imitate intelligent human behavior, it’s often due to AI. The following key properties are characteristic to AI, and they help to differentiate the discipline from other branches of computer science.
- Autonomy: The ability to perform tasks in complex environments without constant guidance from a user.
- Adaptivity: The ability to improve performance by learning from experience.
AI can be far better understood by considering a few application examples. AI is used for search engines, social media, online shopping, and even your favorite mobile apps. It’s embedded into our everyday lives, making the tasks we complete and the entertainment we seek faster, easier, and better — often without us even noticing.
Machine learning (ML) is a part of AI, with its roots in statistics. ML is the study of computer algorithms that improve automatically through both experience and the use of data.
Artificial intelligence (AI) is fairly new to the animal health sector. Over the past decade, however, researchers, scientists, and entrepreneurs have begun to introduce machine learning and AI into veterinary health, with groundbreaking results. Of course, nothing can replace the expertise of a good veterinarian when it comes to diagnosing and treating diseases in animals. What AI does is help streamline your vet’s job, and that gives her more time and freedom to devote to your dog or cat’s well-being. Let’s look at how AI is used in veterinary settings, and how it positively impacts your dog or cat.
HOW DO VETERINARIANS USE AI?
Imaging
Animals always need x-rays, but there are currently not enough veterinary imaging professionals to take and interpret those images. While teaching staff in universities have been hardest hit by the shortage, even lucrative private veterinary practices are struggling to find and keep veterinary radiologists. AI can fill the gap left by these shortages.
The good news is that AI products are available to interpret x-ray images. They involve cloud-based software that uses AI to read x-rays and interpret them quickly and inexpensively; your vet accesses it by signing into a website and uploading images. The results come back almost immediately so your vet can move on with the process of diagnosis and treatment.
AI is extremely well suited to radiology because pictures really do contain a thousand words. X-rays are filled with data, and AI is able to quickly compare previous and current images, prioritize the data, and analyze images. Veterinary radiologists are still needed to read complex images, but
AI can streamline the analysis process, filtering out mundane and uninteresting x-rays so that veterinarians can concentrate on the images that most need attention and the expertise of a trained clinician.
Data analysis
Thanks to continuous improvements in modern medicine, veterinarians are inundated with data from devices, software, and other sources. While more is certainly better, it can also be overwhelming; large datasets are difficult for humans to read. They also might contain irrelevant data or false patterns. Also, when faced with a firehose of data, a human being is likely to miss the big picture. Not so with AI, which never tires, and can sift through large quantities of data to find complex patterns, unprecedented correlations, or small abnormalities humans cannot see.
For example: Consider some recent work done by veterinarians at the University of California, Davis School of Veterinary Medicine, who worked with a computer engineer to develop an algorithm tasked with finding Addison’s disease in dogs. Addison’s is a rare disorder, potentially fatal because it mimics the symptoms of other diseases. This means it’s often misdiagnosed, going undetected and untreated for years. Dogs with Addison’s present with vague symptoms that look like other conditions, such as kidney and intestinal disease.
Normally, when a sick patient first visits the vet, routine blood tests are ordered — a complete blood count and serum biochemical profile. Because Addison’s patients lack critical hormones, their tests often come back with subtle irregularities that are frequently confused with other conditions. The UC Davis team’s algorithm uses AI to analyze blood work data and detect complex patterns unique to Addison’s. The researchers used the test results of 1,000 dogs to train their algorithm to detect the patterns that signal Addison’s. The algorithm functions as an alert system, using information from routine screening tests to flag patients in which Addison’s disease is likely, and inform your veterinarian that further investigation is necessary. It has been 99% effective in diagnosing new patients.
Diagnosis and prediction
When it comes to life-threatening diseases, it is critical
to catch them before they develop. This may sound impossible, but with the right data, vets can make educated predictions about which animals will develop a disease.
Chronic kidney disease (CKD) in cats is a good example — it’s not reversible, and often, by the time it presents, the patient has already suffered kidney damage. It also tends to affect older cats, so by the time a veterinarian catches a case of CKD, the cat’s quality of life is likely to be severely impacted. If the occurrence of CKD can be predicted, however, the patient can be treated before kidney damage occurs, and the cat’s health and quality of life can be dramatically improved.
For example: Researchers recently developed an algorithm to predict CKD before a cat gets sick. It uses AI to predict whether a cat will develop the disease. Trained on Electronic Health Records (EHR) from
20 years of vet visits, the algorithm looked for specific factors that contributed to CKD in more than 100,000 cats across breeds, geographical areas, and ages from one year old to more than 22.
Using this dataset, the team built a recurrent neural network (RNN) that examines blood work for four factors contributing to CKD: creatinine, blood urea nitrogen, urine specific gravity, and age. The RNN was able to predict whether a cat will develop CKD within the next two years with greater than 95% accuracy. The false positives were very low — a huge benefit for vets who have traditionally dealt with CKD as a difficult-to-detect disease. This model can quickly be implemented in hospital practice or diagnostic laboratory software to directly support veterinarians in making clinical decisions regarding sick cats.
It’s important to remember that while AI is excellent at crunching numbers and digesting a large amount of data quickly, it doesn’t do well at some of the tasks at which humans excel. Creativity, problem-solving without a defined training dataset, and of course, bedside manner, are all human skills that AI cannot duplicate. For this reason, AI is an excellent partner to your veterinarian. By taking the pressure of diagnosis, prediction, or data analysis off the vet, AI allows her to really focus on your dog or cat’s health problems, decide on courses of treatment, and make sure he has the best quality of life possible.
Using AI in the Veterinary Field
The veterinary field involves numerous tasks, many of which are cognitive in nature — whether it be processing a clinical history and combining that information with an examination to generate a set of differentials, or interpreting results from an imaging study or set of blood tests. There are so many areas in the day-to-day working life of a veterinarian where AI could make the job of the individual and team easier, efficient, and sometimes more effective.
Ultimately, the power of AI is akin to having the collective mind of some of the best specialists working for you in real-time, making you, the user, better informed and more confident in your role as a veterinarian.
AI & Diagnostics
One such example is the application of AI in complete blood count (CBC) tests. Through AI and machine learning, blood cell identification methods are accelerating and becoming more accurate. Preprogrammed algorithms are used to facilitate the identification of common abnormal patterns within CBC results. Today’s diagnostic equipment uses sophisticated sensors to detect and assess sample information from five dimensions simultaneously. Then, it takes the cell data captured and provides different views of each cell, allowing the cell populations to be separated without the interference of fragments or other cells. This translates to improved cell characterization of red blood cells, white blood cells, and platelets — giving veterinarians additional insights into underlying hematologic abnormalities.
It is not far-fetched to imagine a not-too-distant future in which it will be commonplace for medical diagnostics to be automatically interpreted by AI and a report to be generated in real time, all to then be reviewed and confirmed by the veterinarian. Automation of clinical interpretation tasks such as these offer the promise of freeing up valuable veterinary time and cognitive bandwidth — both of which can then be better spent either seeing more cases or investing more in existing cases, including continued professional development.
There are many examples of exciting applications of AI within the veterinary field, and we are already moving swiftly into an era of AI-enhanced, evidence-based veterinary practices. While challenges and concerns are to be considered, the veterinary field is not immune to the onward march of progress — with AI, in many ways, we are sprinting into the future.
Data plays a key role
A lot of analysis tools already exist that allow data scientists to analyse and visualise data to find trends and patterns, but it is all based on historically collected data. This is where artificial intelligence can intervene as AI has a lot of value in real-time predictive analysis. With the right data collection methodologies, real-time analysis of how a certain disease is spreading can allow an AI to find commonalities among the samples and identify the cause of the disease spread (such as an infected batch of food from a certain facility). In other cases, AI could study the pattern of the disease spread, to predict the geographical regions it will spread to and allowing farmers, veterinarians and pharmaceutical companies to be better prepared for it.
There are further benefits in using ‘smart cameras’ (cameras powered by AI). There has been research in the use of facial recognition with animals and it now proves that smart cameras can identify individual animals. Such smart cameras, mounted at fixed locations or on drones, can monitor the behaviour of animals and collect data on each individually identifiable animal. For example, monitoring the amount of food a specific cow is taking on a dairy farm, correlated with the amount of milk output with that specific cow – can allow the farms to optimise the milk output out of each cow.
Artificial intelligence in animal health
Artificial Intelligence is also bringing changes to business models in many sectors; particularly regarding when human interaction or action is required, and what decisions are automated. Big data is also changing the way managers make business decisions, making them increasingly quantitative decision-makers, relying on data, rather than just on intuition. IBM is already working on ‘AI assistants’ for veterinarians, that can instantly recognise species the moment a pet or animal walks in through the door, and look up through a database of over 800 medical conditions and caners.
Increasing accuracy of automated veterinary diagnostic tools can change when a veterinarian physically interacts with an animal, with AI deciding whether a physical check-up is even warranted at a particular stage, or with particular symptoms. A New Zealand-based app called ‘Betty’, is doing just this, helping farmers decide if their sick cow is an emergency and whether they require a veterinarian. Artificial intelligence could also give the rise of ‘Smart Farms’, which may automatically diagnose a sickness and administer remedies/cures to the affected animal (as part of its feed), without any human involvement.
What the future holds
Artificial Intelligence has already proven itself to be a major driver for increasing efficiency and productivity in other sectors, and it is about time we expand its implementation in animal health. We can already see the rise of AI-powered apps and tools for farmers and veterinarians, but the options are still limited, and still in their development stages. The future of AI in animal health is not yet certain, but it sure is an exciting future to look forward to. What do you think – will AI take over, play a key role or is it just another gimmick?
Uses of AI in veterinary medicine.
Detect Rare Animals’ Diseases
One of AI’s primary uses in veterinary medicine is detecting potential animal diseases. Our fluffy friends cannot talk, which makes it difficult for doctors to diagnose their health conditions. Fortunately, the AI application used in veterinary medicine can help doctors correctly identify more intractable diseases for animals.
For instance, AI can be used to detect Addison’s disease in dogs, which can have very serious consequences for dogs. Addison’s disease is usually hard to diagnose in dogs. Since the symptoms of the disease are similar to many common dog illnesses, this often leads doctors to overlook the possibility of Addison’s disease. A dog with Addison’s disease may have symptoms such as:
- Losing weight
- Unhappy and depressed
- Drinking too much water
- Unable to face stress
- Frequently diarrhea
- Poor appetite
Even though properly timed treatment can help dogs with Addison’s disease have normal lifespans, all these vague and common symptoms have become hindrances for doctors trying to accurately diagnose the disease.
Thankfully, the application of AI provides a solution to this dilemma. Nowadays, an AI-based algorithm developed by the veterinarians at the University of California, Davis School of Veterinary Medicine, plays a significant role in disease diagnosis. Addison’s disease leads to insufficient hormone secretion in dogs, which will show some slight differences in dogs’ blood test results. The AI-based algorithm is trained to identify the differences and report abnormal blood tests.
In this situation, the algorithm is working as an alarm, telling veterinarians which medical cases are suspected of potential Addison’s disease. Then, veterinarians will proceed with further diagnostic testing for these cases. With learning and training, the accuracy rate of this AI-powered algorithm can reach 99 percent.
Predict Potential Animal Disease
In addition to detecting rare diseases, AI applications in veterinary medicine can also help doctors to predict animals’ healthy conditions and the risk of getting potential diseases in the following years and provide proactive care.
For example, AI is largely used in predicting Chronic kidney disease (CKD) for cats. According to AAHA, CKD has become the 1st cause of death in cats older than five, and about 30% of cats over 12 years old suffer from this disease. The usual symptoms of CKD include:
- Bad breath
- Poor hair quality
- Weight loss
- Depression
- Variable appetite
Well, an AI-powered algorithm was developed by the American College of Veterinary Internal Medicine. By analyzing the data from more than 150,000 cats, the algorithm now has the ability to predict the potential risk of a cat developing CKD. By analyzing and learning from the huge amount of health data, the algorithm can predict whether a cat will get the disease in the next two years or not with an accuracy of 95%.
Even though the prediction can not prevent the occurrence of CKD, it allows veterinarians to take proactive care of the “future patients,” which will help the cats to suffer less and live longer and happier lives.
Automatically Code Notes for Doctors
Taking notes is significant and necessary for both doctors and vets. The records provide the health background and medical experience of pets, which work as accurate resources for vets to diagnose the condition of animal patients.
However, taking vet notes is not that simple. Before AI intervention, the traditional way of taking vet clinical notes was by handwriting, which made the work time-consuming, messy, and hard to copy. The difficulty had been bothering vets for decades until James Zou, an assistant professor of biomedical data science, invented an AI-powered algorithm called DeepTag.
Using artificial intelligence and applying natural language processing, the software is able to understand the texts of doctor’s notes and transform the textual information into codes that represent specific symptoms and diseases. In this way, it becomes easier to extract information from clinical databases, compare medical cases, and identify suspicious cases with potential disease risks.
Better Interpretation of Medical Images
Another way AI benefits veterinary medicine is by providing a better interpretation of medical images, such as radiology results. AI-based software can take over the simple and tedious work of veterinary radiologists, such as analyzing data, collecting information, and classifying cases. More importantly, it can provide suggestions to prioritize serious cases based on its interpretation of medical images.
Even though the AI-based software cannot fully replace the doctor’s role, it can streamline the process and improve the efficiency of diagnosis.
How is AI being used in animal health?
The fact that AI is good at repetitive, data-centered, and often mind-numbing tasks that most humans aren’t equipped for means it has the potential to drastically change veterinary medicine. In fact, AI is already being used in various areas of the animal health industry to help veterinarians diagnose, treat, and make better decisions about animal health.
Imaging
Animals always need x-rays, but unfortunately, there are not as many people to take and interpret those images. Recently, there haven’t been enough veterinary imaging professionals.¹ While teaching staff in universities have been hardest hit by the shortage, even lucrative private veterinary practices are struggling to find and keep veterinary radiologists. AI can fill the gap left by these shortages.
At least two startups are offering AI products that interpret x-ray images.² The computer software created by these companies uses AI to read x-rays and interpret them quickly and inexpensively. The software is cloud-based; users access it by signing into a website and uploading images. The results come back almost immediately so the veterinarian can move on with the process of diagnosis and treatment.
AI is extremely well suited to radiology because pictures really do contain a thousand words. X-rays are filled with data, and AI is able to quickly compare previous and current images, prioritize data, and analyze images. Veterinary radiologists are still needed to read complex images, but AI can streamline the analysis process, filtering out mundane and uninteresting x-rays so that human doctors can concentrate on the images that most need attention and the expertise of a trained clinician.
Data analysis
Thanks to continuous improvements in modern medicine, veterinarians are inundated with data from devices, software, and other sources. While more is certainly better, it can also be overwhelming; large datasets are difficult for humans to read. They also might contain irrelevant data or false patterns. Also, when faced with a firehose of data, a human being is likely to miss the big picture. Not so with AI, which never tires, and can sift through large quantities of data to find complex patterns, unprecedented correlations, or small abnormalities humans cannot see.
For example, consider some recent work done by veterinarians at the University of California, Davis School of Veterinary Medicine, who worked with a computer engineer to develop an algorithm tasked with finding Addison’s disease in dogs. Addison’s is a rare disorder, potentially fatal because it mimics the symptoms of other diseases. This means it’s often misdiagnosed, going undetected and untreated for years. Dogs with Addison’s present with vague symptoms that look like other conditions, such as kidney and intestinal disease.
Normally, when a sick patient first visits the vet, routine blood tests are ordered — a complete blood count and serum biochemical profile. Because Addison’s patients lack critical hormones, their tests often come back with subtle irregularities that are frequently confused with other conditions. The UC Davis team’s algorithm uses AI to analyze blood work data and detect complex patterns unique to Addison’s. The researchers used the test results of 1,000 dogs to train their algorithm to detect the patterns that signal Addison’s. The algorithm functions as an alert system, using information from routine screening tests to flag patients in which Addison’s disease is likely, and inform veterinarian that further investigation is necessary. It has been 99% effective in diagnosing new patients.
Diagnosis and prediction
When it comes to life-threatening diseases, it is critical to catch them before they develop. This may sound impossible, but with the right data, vets can make educated predictions about which animals will develop a disease. Chronic kidney disease (CKD) in cats is a good example — it’s not reversible, and often, by the time it presents, the patient has already suffered kidney damage. It’s also a disease that tends to affect older cats, so by the time a veterinarian catches a case o CKD, the cat’s quality of life is likely to be severely impacted. If the disease occurrence can be predicted, however, the patient can be treated before kidney damage occurs, and the cat’s health and quality of life can be dramatically improved.
Researchers recently developed an algorithm to predict CKD before a cat gets sick⁴. It uses AI to predict whether a cat will develop the disease. Trained on Electronic Health Records (EHR) from 20 years of vet visits, the algorithm looked for specific factors that contributed to CKD in more than 100,000 cats across breeds, geographical areas, and ranging in age from one year old to more than 22.
Using this dataset, the team built a recurrent neural network (RNN) that examines blood work for four factors contributing to CKD: creatinine, blood urea nitrogen, urine specific gravity, and age. The RNN was able to predict whether a cat will develop CKD within the next two years with greater than 95% accuracy. The false positives were very low — a huge benefit for vets and pet owners who have traditionally dealt with CKD as a difficult-to-detect disease. This model, say the researchers, can quickly be implemented in hospital practice or diagnostic laboratory software to directly support veterinarians in making clinical decisions regarding sick cats.
Precision medicine
Prediction isn’t just about the disease; it’s also about treatment, because not all patients respond well to the same therapeutics. For instance, the treatment of blood cancers, the most common of canine cancers, can benefit from AI.
In general, chemotherapy is the most widely-used treatment option for canine blood cancers, but finding the right drugs for each patient can be a challenge for vets. A wait-and-see approach after drug administration to a patient’s body can be costly, time-consuming, and take a toll on the patient and the humans who love them. AI can help veterinarians find the most effective drugs for each individual patient and exclude ineffective ones before treatment even begins, often called a “precision medicine” approach.
One application of AI for cancer precision medicine involves the analysis of various drug responses using “live” tumor cells from canine lymphoma patients.⁵ 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.
The study found that patients who had been tested achieved clinical remission much more quickly with their selected drugs. Such precision medicine service enables veterinarians to recommend drugs or drug combinations that will help their patients, rather than taking a trial and error approach to chemotherapy.
AI’s role in the veterinary office
Veterinarians have always applied leading edge advances in technology to animal health: from digital imaging to sophisticated anesthesia, new technology has changed and improved veterinary medicine. So it’s no surprise that vets have begun turning to AI to improve the care and quality of life of their patients.
While AI is excellent at crunching numbers and digesting a large amount of data quickly, however, it doesn’t do well at some of the tasks at which humans excel. Creativity, problem solving without a defined training dataset, and of course, bedside manner, are all human skills that AI cannot duplicate. For this reason, AI is an excellent partner to veterinarians. By taking the pressure of diagnosis, prediction, or data analysis off a vet, AI allows vets to really focus on their patients’ health problems, decide on courses of treatments, and make sure an animal has the best quality of life possible.
How can artificial intelligence change how you practice?
Through AI platforms, computers can learn how to mimic human thought processes and cognitive functions, but with increased speed and learning capacity. Artificial intelligence has several subdisciplines, including machine learning, unsupervised learning, and deep learning. The versatility of AI makes it useful in a variety of medical disciplines. In human medicine, AI has already found applications in areas such as drug design, anesthesiology, cardiology, radiology, oncology, and infectious disease management.
In general veterinary practice, protocols for vaccination, parasite prevention, and many aspects of wellness care are well-established. But, comparable models for treating sick or injured patients are less standardized. AI bridges this gap, by creating protocols and algorithms for the care of sick and injured patients. In doing so, AI brings advanced technology to veterinary practices. It also allows veterinarians to focus more on patient care, by providing reliable, consistent, transparent answers in a timely fashion to clinicians and pet owners.
HISTORY OF AI
Artificial intelligence is the branch of computer science devoted to creating systems to perform tasks that would normally require human intelligence.1 It is a broad umbrella term that encompasses a variety of subfields and techniques. While most current applications of AI have been developed in the last decade, the concepts and ideas have been around for at least 70 years. In the late 40s and early 50s, the first concepts of AI were introduced to the scientific community.Perhaps most famously, British computer scientist Alan Turing was one of the first to introduce the concept of computers performing intelligent tasks in 1950.The term artificial intelligence was coined by John McCarthy in 1955.
Artificial intelligence was heavily researched in the second half of the 20th century but experienced little advancement in its application and scope due to limitations in available computing power. However, with major advances in computer processing power in the past decade and the digitization and availability of large amounts of data, AI has taken off..In medicine, such data includes but is not limited to medical imaging such as radiographs, CT, and MRI; photomicrographs obtained by cytology and histology; and data from medical records including free text and bloodwork results. A major application of AI in medicine is to glean insights from these massive data sets with the aid of computer algorithms to help make or improve diagnoses and improve therapy and patient outcomes. As we will see, the way in which this information can be analyzed and applied varies and depends on the desired outcome of the AI.
Types of AI
Some people picture AI as computers that talk to us or, in the providence of most science fiction, robots aiming to take over the world. However, these types of AI are not what is meant when we refer to AI for medical practice. While there are many ways to classify AI, one classification scheme relates to the scope and ability of the AI.5 Artificial intelligence that has humanlike intelligence is known as artificial general intelligence or strong AI. Artificial intelligence with greater than human intelligence is known as artificial super intelligence. These types of AI generate a lot of fear on the basis of their portrayal in science fiction. However, they are entirely the providence of fiction. Despite the increasing complexity and ability of computers, no known computer systems are near either general or super intelligence. In fact, some researchers argue this type of AI may never exist.5 In reality, AI systems are often designed for a very specific function, such as walking, talking, deciphering, and responding to verbal commands or, in the case of medicine, providing a solution to a specific clinical question.
The type of AI that exists today in medical applications is broadly considered artificial narrow intelligence. That is, these AIs are designed for a specific task and since they only do a specific task, they are considered narrow or weak. In everyday life, examples of artificial narrow intelligence are used in tasks like the spam filter on your email or predictive text for messaging and in word processors. An example in medical practice is the detection of an abnormality on a radiographic image. While these are still complex tasks that require large amounts of data (eg, thousands of radiographic images) for training the AI system, these AIs are still considered narrow or weak when viewed from the level of intelligence.
Methods and Terminology for AI
To understand how AI works and how to implement it in veterinary practice we must first understand some general concepts and terms used in AI. While there are many techniques, which are outside the scope of this article, some general terminology will be encountered by any individual reading about AI. A commonly encountered term is machine learning (ML), which is a subfield of AI in which algorithms are trained to perform tasks by learning patterns from data rather than by explicit programming. It is unusual to think of computer programs as “learning,” but this is exactly what differentiates AI from rule-based computer programs.
Machine learning models can learn in 3 ways: supervised learning, unsupervised learning, and semisupervised learning . In supervised learning, labeled data sets are used to train algorithms to classify data or predict a number.7 Supervised learning requires the outcomes of medical data—that is, the diagnosis or classification—to be known prior to training the ML model to the task (labeled). In this type of ML, the algorithm needs 2 things: lots of data and the corresponding labels. Supervised learning is the most common form of learning for ML algorithms in medical practice as they provide the most clinically relevant data.7 However, unsupervised learning, where the ML algorithm generates its own set of criteria by which to classify data or predict outcomes, can be valuable, especially for large data sets in which the features to distinguish groups are unknown.7 In this type of ML, the algorithm has data but none of it is labeled. The goal of unsupervised learning is to help make sense of the data by means of examining it in detail and seeing if there are relationships or correlations in the data that might be clinically useful. Semisupervised learning uses a combination approach and can be valuable to developing algorithms where some of the data is missing an outcome.
Examples of types of machine learning as they apply to veterinary radiology. In the case of supervised learning (top), training data in the form of radiographic images are labeled in this example into different classifications with respect to the type of study. An artificial intelligence (AI) model that encounters a novel image (input) can then classify the image on the basis of what it has learned from the training data. In the case of unsupervised learning (bottom), data without labels are grouped by the AI model into images with similar features.
There are a number of ML techniques with sophisticated names like support vector machines, decision trees, naïve Bayes, logistic regression, and linear regression that rely on supervised learning. But fundamentally, these are algorithms that simply predict an outcome on the basis of some input data. Unsupervised ML techniques include clustering and principle component analysis. Fundamentally, these algorithms find associations in data and help collapse large data sets into smaller representative ones . These methods are commonly considered classical ML.7 In contrast, modern ML involves the 2 other most common terms in AI: artificial neural networks (ANNs) and deep learning.
Artificial neural networks are computer systems composed of layers of connected nodes.3 They are named as such because they mimic the biological process of neurons, which have many input and output synapses and interconnect with other neurons with many input and output synapses. As this is an in-depth concept, it may be beneficial to consider the example of making a diagnosis of “pulmonary nodules” on a medical image both from the perspective of a human observer and an AI . In the case of the human observer, our eyes receive light signals from the computer monitor that trigger photoreceptors in the eyes. These receptors then trigger several interconnected optic nerves in the retina that send information on the shape and color of the object through the optic nerve bundles and chiasm to the brain for interpretation. Not all neurons in the retina pass signals through the exiting optic nerve to the brain, and only few neurons in the brain decode the neural messages as “medical image.” Many nerves intentionally switch on and off on the basis of the size and color of the object detected. Then, the trained veterinarian interprets this image as “pulmonary nodules” since they have repeatedly seen very similar images. The training and experience of the veterinarian have developed a trained set of neural pathways that combine to generate an output of “pulmonary nodules.” The input is the shape and color of the object, and the output is the brain interpreting the input through a densely connected network of neurons switching on and off uniquely as “pulmonary nodules.”
An illustrative example of image classification by an expert veterinary observer and an artificial neural network (ANN). In the case of a human observer, light photons trigger nerves in the retina, some of which activate and send signals to the brain. Within the brain, networks of neurons are selectively activated (green) or deactivated (red). This in turn triggers a response by a person with training and experience to classify the image as having pulmonary nodules on the basis of its similarity to other images seen previously. In an ANN, pixel data from the images enter at the level of the input layer before progressing through a series of hidden layers. Depending on the degree of activation or deactivation in the hidden layer, the ANN would classify the image, in this case correctly as a patient with pulmonary nodules.
Just as neurons receive sensory input and require a certain level and combination of interconnected activation in producing an output, an ANN consists of a network that converts input data to an output. Input data for the ANN may be the medical image. The image is processed and filtered through a series of hidden layers that help predict the output. In training an ANN, weights are applied to the hidden layers to minimize incorrect predictions. An ANN could involve just 1 hidden layer, but its value lies in the depth of layers. This is where deep learning comes into play. Deep learning occurs on an ANN with typically many more than 10 layers (hence, deep). This allows the algorithm to handle incredibly complex data, like medical images. Artificial intelligence for medical image analysis typically relies on a convolutional neural network, which is the most popular neural network for applications of AI in medicine.
It can be challenging to understand how these terms fit together, and many are often inappropriately used interchangeably. The Russian nesting dolls analogy is one offered by IBM in the IBM Cloud Learn Hub. In this way, the concepts of AI can be thought of as Russian nesting dolls, each fitting inside one another. Machine learning is a subfield of AI. Deep learning is a subfield of ML, and neural networks make up the backbone of deep learning algorithms.
Part of the promise of AI is the opportunity to identify aspects of data that are not immediately apparent to a human observer. Radiomics is a field of study in which an AI algorithm can be used to extract a large amount of quantitative features from medical images.The goal of radiomics is to determine the phenotype of the imaging finding; for example, what kind of tumor is present. One can imagine the implications if a type of mass—for example, a splenic mass—could be determined from the imaging features alone or if precision-based treatments could be developed from noninvasive phenotyping of disorders.
The final important concept for veterinarians to appreciate is natural language processing, which is a subset of AI in which computers can decipher and attribute meaning to text and spoken words. These programs draw from ML techniques and linguistics and will be incredibly important for efficient medical practice. While medical imaging and numerical data can be directly analyzed with ML techniques, some of the most important data in medical records are text based. Natural language processing can play an important role in efficiently extracting information from medical records.
Uses of AI
The potential applications for AI in veterinary medicine are immense. Artificial intelligence can foreseeably be introduced into nearly every aspect of veterinary practice including diagnostics, companion animal care, population medicine, agriculture, research, education, and industry. If digital data exists and can be curated, AI technologies could be leveraged. In this article, our examples and considerations are limited to diagnostic medical imaging. This is due in part to the authors’ backgrounds, but it is worth noting that the use of AI in clinical diagnostic imaging practice is foreseeable within the next few years, largely because much of the data (radiographs, ultrasound, CT, MRI, and nuclear medicine) and their corresponding reports are in digital form.
In veterinary diagnostic imaging, applications of AI focus on the detection, segmentation, or classification of features in the image.In the case of detection, abnormalities can be identified in images; for example, if a pulmonary nodule is present in the image. In the case of segmentation, structures can be delineated within the image; for example, defining the border of the nodule in the image. In classification, the feature or image can be assigned a category; for example, if the patient is positive or negative for metastasis.
Similar to the tasks undertaken by radiologists,detection, segmentation, and classification algorithms can be applied in other veterinary professions. These types of applications can aid in triage to ensure timely care for critical patients or timely radiology reports for images in which an AI has identified an abnormality. If AI systems can connect to digital health informatics and patient records systems, they can be used to streamline processes to reduce the workload on veterinary professionals. Or, if used in conjunction with diagnostic equipment, they can assist veterinarians in making accurate diagnoses on the basis of either the detection or classification of disease.
To date, there are a few commercially available AI systems for veterinary diagnostic imaging; however, these systems have not undergone the rigors of peer review. A majority of the peer-reviewed AI applications for veterinary imaging to date focus on proof of concept to show that AI can accurately detect abnormalities in the canine thorax.
Considerations and Challenges
As AI becomes more available and accessible in veterinary medicine, veterinarians should consider a number of important challenges that will be encountered. We propose the following considerations for veterinarians as they consider the adoption of AI into veterinary practice.
Use cases
For AI to be most effective, it requires a directed purpose. Determining to which aspects of practice AI should be applied is a job for veterinarians. Veterinarians provide a level of expertise that determines which use cases are most valuable for the profession. Questions to consider include the purpose of the technology and how it benefits the welfare of the animal or the profession. Without veterinarians taking an active role in asking these questions when they are offered an AI-based solution, there is a risk of prioritizing profits over clinical outcomes and the well-being of both veterinary professionals and their patients. The goal of AI should be to improve veterinary practice, animal health outcomes, patient quality of life, and the lives of veterinarians. This requires well-thought-out use cases and active veterinary stakeholders.
Good data
Data is not cheap, and AI developers know this. In an era in which AI becomes a commercially viable entity, data will become like a currency. Who owns the data, how it will be used and managed, and eventually how it will be used in an AI model will be shared and should be well established at the outset of AI projects. Universities and academic institutions often have the benefit of local resources that can provide guidance on data sharing, licensing, and intellectual property; however, this may not be the case for smaller institutions and clinics. We encourage veterinary professionals to recognize their role in curating data and maintain ownership. Data must be labeled in most cases, representing a role for veterinarians that should not be underplayed.
Data is only as good as the individual labeling it and the label they apply. Labels must have an underlying ground truth. This raises the concern of what determines the truth of the label, particularly in medicine when there are often many confounding factors associated with a diagnosis or treatment. Additionally, there are differences in opinions between veterinarians or between institutions. For example, radiologists differ widely in their sensitivity to diagnosing small animal patients with a bronchial pattern. Within this range of diagnosis is a second range relating to the clinical significance of the finding. The subjective nature of interpreting medical tests creates variability in data labeling, which can lead to over- or underfitting of the AI model.
Bias is an additional consideration of concern in AI. Bias may arise from training sets that have skewed breed or geographic distributions or imposed by the method of curating the ground truth. In the case of veterinary radiology, patient positioning and radiograph quality will play a significant role in the performance and bias of algorithms. For example, AI trained on perfectly positioned and exposed radiographs may not perform well with radiographs acquired with suboptimal technique. While it is not realistic to assume data sets are perfectly labeled, careful review of data before attempting to use it in AI helps establish implicit assumptions in the model.
Not all data is useful data. It is also important to establish what data is worth properly labeling for AI use. Big data approaches in health care can be data rich but information poor (also referred to as DRIP). Indeed, some of the biggest health-care challenges are not related to what problems can be solved with AI but creating and curating good and clean data for AI.
Open data
As researchers, we believe in accessible and equitable data sharing. While there are many reasons why one should consider accessible and equitable data sharing, some are as follows. First, having access to transparent data permits an opportunity to validate data by others. Data sharing permits institutions and researchers to identify potential biases in their data collection processes before developing an AI algorithm. Second, open access allows an opportunity to validate that the data and AI system can be used on different platforms (interoperability). Data, and any AI developed from it, should not behave differently for PC, MAC, or Linux users. Third, accessible data allow different researchers and institutions to formulate different and potentially better AI solutions. Last, data sets in veterinary practice are much smaller than those found in human medicine. Combining data from different institutions allows the opportunity to increase the diversity of data sets for ML, which can eventually lead to more robust and generalizable models. It is our opinion that open data is good for veterinary medicine.
This promotes the ideal goals of AI rather than monetization. However, it must be acknowledged that there is a cost associated with data storage and sharing. This cost is likely to fall on corporations and organizations that will likely retain at least some of the rights to the data. Partnerships between academic institutions, granting agencies, and professional organizations may be necessary to facilitate open, curated, and good veterinary data for the community.
Ethics and regulation
There are many ethical questions that arise with AI in veterinary medicine. Initially, as discussed above, a question exists as to who owns the data. This is important to consider as data in the world of AI has immense value. It is also important to consider confidentiality and security as it relates to patient information.
While AI products for medical applications in humans are regulated by the FDA in the US or other similar bodies worldwide, there is no regulatory framework for AI in veterinary medicine. Regulation will be critical for the success of AI in veterinary medicine to encourage ethical, responsible, and directed use.
Implementation
While there is much promise behind AI, implementation to use AI most effectively in practice has proven to be a challenge in the human medical field. While hurdles such as FDA approval are not currently in place for veterinary medicine, the other challenges to implementation will be similar. Good AI governance practices in veterinary medicine should be developed if these technologies are routinely integrated into practice. Some critical responsibilities include establishing use cases, fiscal responsibility when AI technologies are purchased, sufficient infrastructure (software or hardware) and resource support, establishing good data management principles, thorough acceptance testing and clinical deployment including training and education, and comprehensive quality management of these systems throughout the technology’s life cycle.
Client acceptance
Another aspect not to be overlooked is how clients respond to AI technologies. While somewhat outside the control of veterinarians, client acceptance of AI will be vital for its success. Veterinary professionals can promote acceptance with transparency and client education.
Since animals do understand expressions, body language, voice modulations and have memory for the experience they share with the humans; all these factors can aid the making of AI-based software and solutions that improvise pet care and save wild animals from being extinct. The importance of artificial intelligence is increasing to the extent that we now search for the contributors who can revolutionize the machine learning to facilitate the coverage of various fields and be of human assistance.
ARPITA MUKHERJEE,IITB
REFERENCE-ON REQUEST
IMAGE-CREDIT GOOGLE
