Ever wonder what happens when a sample is sent to UF’s Diagnostic Laboratory? Dr. Michael Dark, clinical associate professor and director of CVM’s veterinary diagnostic labs, takes listeners behind the scenes of a diagnostic laboratory to explore the journey of a sample from collection to final result. He’ll detail the process, people and technology that ensure every specimen is accurately processed, analyzed and interpreted, turning samples into actionable information that helps clinicians and clients make informed decisions and support better patient care.
DANA HILL: Welcome to Animal Airwaves Live here on WUFT-FM, our weekly hour long show devoted to the discussion of the health and welfare of animals. So happy you could tune in today on this Friday, September 4, 2026. And happy to welcome to the program my guest from the University of Florida College of Veterinary Medicine, Dr. Michael Dark. And the topic today, Tracking the Tube: A Day in the Life of a Diagnostic Sample, From Collection to Conclusion.
DH: That includes a lot of different kinds of samples and we’re going to find out what those are today. So, first of all, let me welcome you back to the program. Dr. Dark, really glad that you could be here.
MICHAEL DARK: Thanks, Dana. Thanks for having me.
DH: When we think about samples, we think perhaps about the – what happens when we take our pets to the veterinarian’s office and the veterinarian will do an exam, look over our pet, you know, look inside their mouths and maybe, you know, touch their tummies and that sort of thing. And then there might be an occasion which the veterinarian says, okay, we’re gonna just take the pet to the back and we’re gonna get a sample or two. This happens in a pretty good number of veterinary visits, would you say?
MD: Yeah, yeah, I’d say a lot of visits end up getting samples, whether it’s kind of routine visits, but especially in animals who have some kind of problem where they’re getting worked up.
DH: And would you say then that these samples for, you know, I mean, for in, well, visits and in visits where we suspect there may be a problem. And with these different kinds of samples, though, we’re talking about everything from, I don’t know, fluids like urine or something like that, to blood to maybe even tissue samples.
MD: Yeah, and a lot of that depends on exactly what’s going on with the patient. But it’s pretty common to look at blood to try and get an idea about a lot of different conditions. It’s not uncommon for vets to collect stool samples to check for parasites. And then if there’s a mass or an issue, then tissue samples can tell you quite a bit about what’s going on. Because we can look at those under the microscope and get a better idea of what the disease is.
DH: Then maybe let’s kind of start with some of the types of samples then, broadly speaking, that veterinarians might be looking at and what circumstances might lead them to collect these samples. You mentioned blood. That’s got to be one of the more common samples, correct?
MD: It is, yeah. And for instance, a test a lot of people are familiar with heartworm testing in dogs is a huge thing, especially here in Florida, where we have mosquitoes nearly year round. Veterinarians will take blood samples a lot of times now, heartworm testing can be done in clinic, so they can take a sample of blood and apply it to a kit that’s a little like the Covid test kits a lot of us used during the pandemic to check for Covid. Instead of nasal swab, you put a drop of blood on and it will give you an idea about whether the animal is infected with heartworm or not. And so that’s a rapid test that can be done quickly in the clinic and a result can come back.
DH: If you don’t mind me kind of getting into something complicated right off the bat. How are these sorts of tests devised? Who comes up with the technology that allows you to drop a little tiny sample of blood into a tube or something like that and figure out whether or not an animal might be infected with, say, heartworms or something?
MD: It’s a lot of different people and groups and disciplines. It’s one of the things that I actually enjoy a lot about working at a place like UF where it’s not just veterinarians. The whole campus has experts in a lot of different areas. So, taking a test like a heartworm test, it’s people who can design antibodies to detect that organism. It’s people who can design the fluid handling; it’s people who can convert the test into a little tiny package.
MD: So there’s a lot of folks involved to develop a new test. A lot of the mechanics of testing like the heartworm test are fairly well known. So, a lot of tests, if you’re developing a new test, you’re taking currently developed packaging and mechanics, and you’re adapting it to, say, a new disease process. And so a lot of that would be more microbiologists or virologists or individuals who are studying that specific disease, finding something that lets you identify that disease and tell it apart from other diseases that are similar, and then developing that into the technology of an existing test to package that up.
DH: And then if this is a test that seems like it can be useful to a pretty large number of veterinarians or other researchers, perhaps it becomes commercially available. Some medical company creates these in volume and sells them to clinics across the country and the world.
MD: Yes. And a lot of veterinary schools, including UF, have developed the initial test and then through patent licensing, will license it to companies to then produce the test in volume and sell that out at scale.
DH: Yeah. Now, I don’t mean to get too much further down this rabbit hole. But when we think about diagnostic testing for human disease, these tests must be approved, probably by the FDA. Does the same thing hold true for tests for veterinary diseases?
MD: Not broadly. The difference is a lot of the requirements are voluntary. So, for instance, UF’s Diagnostic Lab, as well as a lot of other labs across the country, are accredited by a group called the American Association of Veterinary Laboratory Diagnosticians. So that’s a voluntary process. We sought out accreditation.
MD: But for that group, if you are performing a test diagnostically on patients, you have to be able to show data that says that the test works the way you say it does, that you understand what the sensitivity and specificity of the test are and that it’s reproducible, and that you’ve – everybody who does the test has been trained in how to run it and interpret it successfully.
DH: Yeah. You know, the average pet owner who comes in isn’t going to be aware of the accreditation that you all have sought out and received, but it certainly matters. And if it didn’t, you all wouldn’t have pursued this. Talk about a little bit about why you would seek to voluntarily pursue a certification and what that kind of means to y’ all who do this sort of work.
MD: We sought out accreditation for several different reasons. Part of it is our lab and the folks who work in it genuinely have a dedication to quality. We are all pet owners. We all understand what these results mean in terms of the health of the animals that are having the tests done. And so the lab staff has a real dedication to quality.
MD: And so this certification is important to them because it serves as a statement of the importance we put on producing quality, reliable results. The other reason for it is it lets us work more cooperatively with the state diagnostic lab and some of the state agencies and participate a little more broadly in animal health in general, because we can show that, yes, not only are we producing quality results because we’re a veterinary institution and a vet school, but that we’ve had external people come through and verify that we are doing what we say we do. And so it was important for a lot of reasons, but those are the main ones. And I will say that our staff really drove accreditation because of the importance that they put on it. They really felt like it was important.
MD: It was an important statement for us to make of where we think the quality is and where it needs to.
DH: Be when we go backwards and resume discussing, say, a blood sample. I recall a recent visit I had to my doctor’s office when I was asked to give a blood sample. And I’ve done this who knows how many times in my life. I think many of us have. And when the phlebotomist, or whoever it was, did the blood draw, I was reminded of how difficult it can sometimes be for even experienced texts to find a suitable vein and draw that blood.
DH: And how I’ve given up allowing them to take the sample from the sort of inside of my elbow, the inside of my arm there in the crook of my arm, and I just point instead to the top of my hand where I have this like bulging vein that goes across and it’s so easy to see. And it occurs to me that drawing blood from an animal might present some challenges, not least of which is a bunch of fur in the way, maybe skin that’s not quite as easy to see through and identify a vein or something at any given point. How challenging is it to sometimes draw blood from, from patients? Especially when you’re talking about so many different kinds of animals. It’s not just dogs and cats, but you can need a blood sample from anything from a bird to a horse.
MD: The level of challenge is really going to depend a lot on the animal and the skill level of the person drawing it. I have been lucky through my career to have a lot of very talented technicians who were really good at drawing blood. And the folks at the UF hospitals are excellent technicians. They do an amazing job. I don’t know that I would be very good at it anymore.
MD: It’s been a while since I’ve had to draw blood on anything that’s not my own patients. And my wife does a lot more clinical work than I do. So usually, I hold and have her draw the blood nowadays. But yeah, for very small animals it can be really difficult. You have to use small size needles, much like in infants.
MD: You’re going to use a much smaller needle than you would for you or I. And so the, a lot of it is experience. So, as you gain experience in species you work with a lot, it becomes easier to get blood samples. And you also learn some tricks in different species to make it a little easier. And so it’s a lot harder in a species you don’t normally work in, because you’re kind of going based on your experience and what the books say, and then you kind of have to try it a little bit.
DH: I think many of us are familiar with the smallish sort of vial that is the receptacle for the blood that we give when we need lab work done. But of course, animals being such A wide range of sizes from, you know, like we were saying, a bird that might weigh only a few ounces to dogs that can weigh over 100 pounds, or horses that can weigh many hundreds of pounds, cattle and so forth. How much of a sample do you need of something like blood when you’re trying to diagnose some sort of illness?
MD: We can get by with pretty small samples. Usually, we’re talking about a milliliter or two of blood, which is…I mean, if you think about maybe taking a regular sized glass, if you just get liquid on the bottom, you’re probably looking at a milliliter or so. So it doesn’t take very much blood at all. Ideally we get more because it gives us the chance to repeat blood work.
MD: If there’s an issue or if somebody decides they want to add another test on, then the more blood that we have, the more opportunities we have to do additional things. But we can get by with very small amounts of blood. And luckily, as technology has advanced, the amount of blood we need keeps decreasing because we’re using a lot of times for a routine chemistry. For instance, you’re looking at chemical reactions to detect various things in blood.
MD: The drive has been to use less and less blood over the years. And so the chemistry is still the chemistry. The reactions happen as long as you have enough to react. And so a lot of that can happen with very small volumes of blood. And so when we get exotic species in particular, like birds, especially small birds, small rodents, there’s just no way to get as much blood as you would even out of a cat.
MD: And so we have very small blood tubes and can work with fairly small volumes of blood.
DH: When you get the blood into a vial or whatever sort of receptacle, what does it take to, I don’t know, keep that from going bad, if I don’t know the proper terminology. But I would think that, you know, blood would tend to want to coagulate or, you know, who knows what else.
MD: Yeah, a lot of that depends on what tests you’re running. Sometimes we want the blood to clot because we want the liquid part of the blood and not the cells. And so once the blood clots, we can spin it in a centrifuge to get that liquid off, and then we don’t need the rest of it. Sometimes we want it to have all the cells in the liquid. And for that we use a tube that contains an anticoagulant that stops the blood from clotting.
MD: The most common one we usually use is chemical called EDTA. There’s a few others as well. But a lot of your listeners may have heard of heparin. It’s a fairly common drug. And so we have tubes with heparin in them that can stop the blood from clotting as well.
DH: Then with something like blood, if you need to divide it up or retest it later, I mean, does it. Does it store. Is it possible to store it?
MD: That, again, depends a lot on what test you’re doing. So, for instance, if you are looking for the amount of ammonia in the blood, ammonia goes away very quickly. So it gets drawn into a special tube, it gets put on ice and brought to the lab immediately because we have to test it within somewhere around 30 minutes or else it’s not accurate anymore. For a lot of tests, they’re good for quite a while, especially if processed correctly. So if you’re measuring glucose in the blood, you can measure glucose in the blood after a day, as long as you’ve spun the blood down and gotten the cells out.
MD: Because if you leave the cells in, the cells will use up glucose and the glucose will decrease over time. So a lot of that depends on the tests. And I have to say, my specialty is not clinical pathology, who do a lot of the blood testing, but the clinical pathologists are really good about being able to say this test would be good under this condition. And so, generally, we refrigerate the samples that we get to let them be good for longer. And then it comes down to what tests people want to run as to whether that sample is still okay for testing or whether we need a new one.
DH: Now, for everyone who is enjoying lunch at this hour, I apologize, but here I want to talk about some types of samples that I don’t know may be more or less challenging to collect. And I’m talking about, say, urine or stool. And it occurs to me that you could probably without a huge amount of difficulty, get many dogs to provide you a urine sample, let’s say. But a cat, I don’t know, that may be a little bit harder to get a voluntary sample. And I couldn’t even begin to tell you about every other kind of animal.
DH: How do you go about getting these?
MD: There’s a few ways to go around it. For cats, especially, if there’s not multiple cats in a house, there’s plastic litter you can substitute in their litter box so that it doesn’t absorb, so that when they use the litter box, you can then pour the urine out into a cup and collect it that way for all animals, or I guess most animals. There’s also the ability to do what we call a cystocentesis. So you can use a needle and syringe and insert the needle from the outside into the urinary bladder and get urine out that way. Stool samples are usually collected by waiting until the animal defecates and then collecting it.
MD: And for most things, that works pretty well, especially for parasites, we try not to get it off of grass because parasites can get into the sample from the grass. But there are ways around that as well. But those are the usual ways. The other way to get urine, you can also use a catheter, you know, similar to people who have a urinary catheter. I would say that’s less common.
MD: It’s a lot more common to get free catch, which is, you know, you grab a sample while they’re urinating or cystocentesis. And there’s pros and cons to both of those. Especially if you’re looking for bacteria or some sort of microbiology. Cystocentesis is better because it is the least likely to be contaminated by outside bacteria. But free catch is obviously less traumatic for the animal and it’s a lot easier to grab.
DH: Yeah, that makes a lot of sense. Okay, so what about some kinds of tissue samples that you might need to collect? Some of the obvious ones I would think of is maybe a needle aspirate or something if you suspect an animal might have some sort of tumor or something like that. But there’s gotta be others.
MD: Sure. And I would say that a lot of what my specialty is anatomic pathology. So, it’s looking at pieces of tissue under the microscope is one of the big things that we do. Clinical pathology are the folks who look at aspirates from masses, and so we kind of work together.
MD: So it’s pretty common for people to take cells out of a mass by using a needle and syringe and looking at those under the microscope, you can use a slightly larger needle and get what we call a core biopsy. So you put a large needle into a mass, and that lets you take out a piece of tissue without doing actual surgery. But by far and away, the most common way anatomic pathologists get pieces of tissue to look at would be doing surgery. So a lot of those are skin masses or if there’s a mass inside the animal. It’s not uncommon for us to get spleens.
MD: If an animal needs a limb amputation, we might get a limb to look at to see what’s Going on inside there and be able to look at that under the microscope.
DH: Okay. I mean, so some of that sounds like it comes with its own challenges, for sure. But when you are doing something like, probably in your line of work, I mean, now we’re talking about surgery. So this means things like anesthesia and so forth.
MD: Yes. And so a lot of recent advances in surgery have been towards things that make that less invasive. So UF has folks who are specialists in minimally invasive surgery. So they use things like endoscopes, like they use in people where they can put a camera into a small hole instead of opening the whole animal up to be able to get samples out to find out what’s going on. Or instead of opening the animal up and taking pieces of intestine, they’ll use an endoscope down through the mouth to be able to grab samples from the inside to let us see what’s going on.
MD: And so, there’s a variety of different ways to get that. They generally, in animals, all involve anesthesia. It’s. It’s not like people where you can just use a sedative, and as long as they’re not moving around too much, you’re okay. It’s a little more work in animals to get those.
DH: But there is a significant benefit. For instance, when you’re taking a larger sample, if there is some sort of illness or disease that you’re hoping to uncover or reveal, there may just be no substitute for the larger kind of samples.
MD: Sure. And the other aspect of that is sometimes that surgery is curative as well. If we can get a large enough sample out and remove the disease, that may not be a problem going forward either. And part of my job is to help surgeons figure out, did they get everything in the surgery, how invasive does this look? And so those samples give us, having a larger tissue sample gets us a much better idea of that than smaller ones.
DH: So before we take our first break, then, Dr. Dark, I wonder if you can tell me what are some of the most challenging samples to collect when we talk about any of the kind of diagnostic testing that our animals might need.
MD: Hmm. I would have to say, over the course of my career, I once had to collect some samples from elephant feet. Someone sent in some elephant feet looking at disease because they were trying to see if they could adjust things. And just the thickness of the skin on an elephant is tremendous. And so that.
MD: That was one of the more challenging ones. I would say, again, from my perspective, is as a pathologist. So the other aspect of our job is doing postmortems. But getting out spinal cords is fairly difficult because they’re encased in a lot of bone. But those would be the big ones.
MD: And then things from very tiny animals. The smaller the animal then the more challenging it is to get samples, just because you’re dealing with much tighter tolerances. So it’s hard to get samples out. I’ve always been impressed. One of the research projects my wife did when she was a vet student was looking at aortas in mice.
MD: And so they would cut aortas from mice into little strips. And that was just extremely small, fine work. And so, there are a lot of people who are very good at doing that sort of thing.
DH: Yeah. But before we take our first break then, Dr. Dark, for listeners who are tuning in and they’re hearing about this different type of sample collection, I think that one thing that we, we ought to all understand, those of us who have pets, and again, some of this happens behind the scenes, they take our pet in the back and then a sample is collected. You know, this isn’t veterinarians aren’t just asking for this stuff for fun. I mean, this is really about diagnosing and treating disease.
MD: Sure. It’s also about what is the least invasive, least troublesome thing to the animal to be able to get the answers we need. And so a lot of times drawing blood, as we’ve all had, it’s an irritation, but it’s not a huge issue. But it can get a tremendous amount of information about what’s going on with a lot of different organs very quickly.
DH: Right. All right. Well, this makes a good place to start or to begin our first break. I want to remind listeners that this is Animal Airwaves Live here on WUFT-FM. I’m Dana Hill.
DH: My guest from the University of Florida College of Veterinary Medicine is Dr. Michael Dirk. And we’re talking about Tracking the Tube: A Day in the Life of a Diagnostic Sample, from Collection to Conclusion. Stay with us. We’ll be back with more right after this. All right.
DH: Welcome back to Animal Airwaves Live here on WUFT-FM, our weekly hour long show devoted to the discussion of the health and welfare of animals. I’m Dana Hill. My guest from the UF College of Veterinary Medicine is Dr. Michael Dark, who’s the clinical associate professor and director of the UF College of Veterinary Medicine’s Veterinary Diagnostic Labs. Dr. Dark, when we left off, we’d been talking about some of the commonly collected kind of samples that you all get.
DH: But it occurs to me that, ultimately, since what you’re trying to do is diagnose and treat disease and illness in animals, often pets, that there are common diseases, common ailments that you guys see on an everyday basis. But then there may be some that are more unusual, not seen perhaps in one geographic region more than, say, another geographic region. And I wonder whether or not there are diagnostic tests that the University of Florida College of Veterinary Medicine doesn’t do, in addition to the many ones, many tests that you guys probably do that maybe other places do not.
MD: Sure, there are definitely labs. We had talked about developing tests earlier. Lots of labs have developed tests that are specific, or they are the lab that found that organism and first developed the test. And so there are some tests where there’s only a lab in the country that runs it. There are other tests that are running pretty much everywhere.
MD: And so we do send out quite a few tests, either because we don’t see that condition a lot, or there’s a lab that’s set up who are the experts in it, and it wouldn’t necessarily be cost effective for us to run it. We don’t see enough of those cases or whatever. And there are tests where we are the experts, where we get samples from all across the country to go ahead and look at that. One of our pathologists, Dr. Hoffman, is an expert in fungal identification, and so she gets fungal samples from all over to determine what the fungus is and give recommendations on that. So, yeah, the veterinary community is very small, and so we work fairly well with each other.
MD: And so we get together and if we have a piece of equipment break, we have labs that we can send samples to while we’re waiting for that to get fixed, if we need to, to be able to get the testing done in a timely manner.
DH: So when I think about some diseases that might be a bit unusual but are significant, I think about something like, say, eastern equine encephalitis. I don’t mean to put you on the spot here, but is there a test that y’ all could do for something like that?
MD: There are, and we’ve typically, when we’re looking at eastern equine encephalitis, we usually call it Triple E, just for short. That is usually in an animal that’s showing neurologic signs. And so one of the other things on our list would be rabies as a possible disease. So commonly for that, we will take samples and send them to the Florida Department of Health, and their lab will test for rabies and then also test for eastern equine encephalitis as well as some of the other encephalitis, West Nile and western equine encephalitis.
DH: Yeah. Okay, so this gets me to another question. So thank you for mentioning that. You know, there are some diseases or illnesses or ailments or what have you that are certainly bad and you wouldn’t want your pet to have them. But then there are diseases that are of concern to…
DH: …that matter to public health overall. And I think, you know, something like Triple E or rabies or something like that is a significant issue. Is it? There’s a different way of handling those sorts of cases.
MD: Yes and no. We call those diseases that can get transmitted from animals to people are called zoonoses. And so we approach cases from the standpoint of looking at what is the risk, first of all, to our lab staff. And so we’re always thinking about, could this animal have a zoonotic disease? Is that something that we need to think about?
MD: And so there is a risk there. So if we consider that that risk is significant enough, we put on protective equipment or handle it in a safety cabinet. We take steps to ensure that it’s handled as safely as possible. And then for rabies, in particular, in addition to animals being able to be vaccinated against rabies, people are able to be as well. And so folks that are working with cases that could have rabies get rabies vaccines before they do that, to make sure that the risk is as low as possible and that if we do get a rabies case, that they’ve already had rabies vaccines and they don’t need to go through the whole series just straight off.
DH: Yeah. Because as many of my listeners will know, rabies is pretty much uniformly fatal if you acquire it. Right. Yes.
MD: There have been less than a handful of human cases that have been documented where a human has gotten rabies and survived it.
DH: Yeah. So, you know, that seems like it’s something that we’d want to make sure that everyone is protected as possible, though, you know, in the general public, like you or I. Well, maybe not you, but me, I probably will never, knock on wood, need a rabies vaccine in my whole life, you know, but folks working who may have a reason to be around animals that could be infected would certainly want to protect themselves. There is no doubt a variety of diagnostic testing that is available for, you know, diseases that are seen on a common, everyday basis. You know, when we talk about diseases that are a little bit more unusual.
DH: Well, first of all, you’ve got to have a veterinarian that maybe…how does it start? With a suspicion of some kind of problem. Right. I mean, an animal comes in, it’s experiencing some kind of symptoms.
DH: These symptoms may be vague or very specific, and there’s a process of elimination that goes on. And then it’s going to take an experienced clinician or veterinarian or so forth to maybe posit, I wonder if this could be X. And then where does it go from there? I mean, then you start to look into what the possible samples that could be collected might be. Right?
DH: I mean, especially if we’re talking about something unusual. You know, you guys certainly there’s plenty of people there with a lot of experience. You know, what are the first steps there?
MD: That drawing on that experience is a lot of where that first step is. We are lucky enough to have a broad enough group of people that there’s not many things that somebody hasn’t seen. At least one of the … to get some kind of start as to, okay, what if this is a possibility? Where would we go with it?
MD: Or if not, know somebody who does. So, as I said before, it’s a very small profession. So I’ve called colleagues at other universities to say, hey, where would you send this test? Or what’s the best way to test for this disease?
MD: Because we don’t do it that often to get a better sense of where to go with that. And sometimes the thing that makes you think of an unusual or rare disease might be you do other laboratory testing and it seems unusual, but the pattern makes you think, or you get an unusual pattern of results. And so you look in the books, and the books say, well, this might be one possibility. Unfortunately, there’s some diseases where there’s no good test for it. It’s what we call a diagnosis of exclusion.
MD: So you rule out a lot of other things by doing a lot of other testing, and you kind of end up with, this is the only thing that fits what the animal is presenting with that we can’t test for. So it seems like that is the most likely explanation.
DH: That brings me to the topic of cost. Let us say, for instance, the average well visit where maybe a sample of one type or another is collected just, you know, to make sure everything’s looking good, kidney values or whatever, I don’t know. But if there is an animal that is presenting with some vague or mysterious kind of symptoms or, or maybe they’re symptoms that could be any number of different things, and you begin to kind of go down this path of doing some diagnostic testing and you collect samples and so forth, and you’re not finding any of the obvious culprits, the number of tests that might be involved could climb and with it, the cost associated with this sort of diagnostic process. There must be some tests that are pretty affordable and other tests that must be rare and expensive.
MD: Sure. In general, I would say the more frequently a test is done, usually the less expensive it is. For instance, heartworm tests are, relatively speaking, I mean, I know actually paying for it is always a little painful, but relatively speaking, it’s not very expensive because they’re done very, very frequently. I mean, every dog should be heartworm tested at least once a year so that in Florida at least, so that is one example. Whereas if you’re doing a test that’s only offered by one lab in the country and they maybe they only do it 50 times a year, it costs a lot to maintain equipment and have staff that are trained in it.
MD: And so the cost of whatever needs to happen for that test has to get spread out over those 50 tests in a year, which makes it much more expensive.
DH: All right, this is where we’re going to take another break, but I want to remind listeners that this is Animal Airwaves Live here on WUFT-FM. I’m Dana Hill, and I’m speaking with Dr. Michael Dark, and we’re talking about Tracking the Tube: A Day in the Life of a Diagnostic Sample from Collection to Conclusion. We’re going to be back in a minute and talk some more about the tube kind of technology involved in this. Stay tuned. Welcome back to Animal Airwaves Live here on WUFT-FM.
DH: I’m Dana Hill, and my guest today from the UF College of Veterinary Medicine is Dr. Michael Dark, and we’re talking about tracking diagnostic samples from collection to conclusion and what it takes to diagnose and treat disease in our pets. And Dr. Dark, I want to talk about some of the technology that you all employ over there. Of course, there’s things like needles and test tubes and that kind of thing. I imagine that there are some pretty sophisticated microscopes and probably some advanced computers and these sorts of things.
DH: What is some of the machinery that you and your colleagues use on a regular basis?
MD: We have a whole variety of equipment in the labs, from machines that take blood and hit it with a laser to analyze what cells are there to determine how many white blood cells and red blood cells and what the size and shape of the cells are. There are machines that do chemical reactions on blood to determine how the liver and kidneys are functioning in the microbiology lab. There are machines that look at a variety of different samples of bacteria to determine whether they’re sensitive or resistant to different antibiotics. We have machines that amplify DNA so we can look for DNA from infectious organisms. We have a machine that actually takes microscopic slides that we’ve traditionally used and will scan it into a computer so you can have a virtual representation of the microscopic slide that you can look at in the computer. We use that now.
MD: We’re using that more and more for meetings with other universities or laboratories or other pathology groups. Because now, instead of having to have multiple sections of a slide cut or have multiple slides cut that you have to mail to someone, you can just email them a link to a file that’s enormous, but then they can look at it on their computer within an hour rather than having to wait a week for them to get slides.
DH: I mean, Dr. Dark, it sounds like that when I’m hearing you describe a lot of this technology, I am kind of dollar signs are flashing in my eyes because not only is the technology itself probably quite costly, but of course, the investment in training and time and so forth that that everyone involved needs to have in order to effectively use this technology, which can come with tremendous improvements in health outcomes. Right. But this is kind of where we are on the cutting edge of, like, veterinary medicine, where, in order to treat and diagnose and treat different diseases, the technology available is ever more sophisticated.
MD: It is. All of us are fairly cognizant because, again, all of us in the lab are pet owners. We all have pets. We all take our pets to the vet. We’re cognizant of the costs involved as well.
MD: And so one of the things that I would say all of us at UF, but the folks in the lab as well are aware of, is trying to make sure that options always exist for people, that we provide a spectrum of care so that you don’t have to be insanely wealthy to get care for your animal. There are tests we may not be able to do, but we want to make sure that there are always options that we can provide to make sure that we don’t price people out of the market as much as possible.
DH: This is – maybe it’ll sound like an odd question, so I apologize in advance if it is – but in this day and age where many people first turn to the internet for questions that they might have about their own health or their pet’s health, it may enter someone’s mind that they need to get their pet some specific kind of test. And I imagine it might happen from time to time that veterinarians might hear that and think that’s probably not necessary. How do you go about talking to pet owners and maybe help guide them away from needlessly spending money on tests that may not be at all related to the suspected disease?
MD: A lot of that is where people’s veterinarians come in. Because I think it’s very important for people to have a good working relationship with their vet and for their vets to have a good working relationship with the labs or lab or labs that they use. Because the veterinarian, if you have a veterinarian you have a relationship with, you have that trust that when you say, hey, I’m concerned about this, what about running this test? If they say, I don’t think that’s justified because of this or that reason, that you have a trust that they’re looking out for you and your pet’s best interests. And so I think that’s crucially important.
MD: And then the veterinarians having the ability to call us and say, I have this patient. I’m concerned about this. What’s the best sample to take? What’s the best way to go about this?
MD: We’re happy to get those calls. We love getting those calls in advance because it helps us make sure that we get the best possible sample and that we can then give them the best answer we can in the most rapid and effective way possible.
DH: Finally, Dr. Dark, everyone who pays any attention to the news now is aware that artificial intelligence AI is potentially a major disruptor for many types of work. And when I say disruptor, I mean it could be for good or bad, let’s say. Do you see a role for AI in kind of diagnosis of diseases? But I’m talking about, like, in analyzing samples or in some other way and maybe augmenting the expertise of a veterinarian.
MD: I think that is where we’ll probably be going for quite a while in veterinary medicine. Things like – one of the things that we do, if we get a tumor that we’re looking at under the microscope – one of the things that is most predictive of what that tumor is going to do is how many cells are dividing and how quickly are those cells dividing. And we look for something called the mitotic figure, which is the DNA of the cell actually splitting into two parts. And when they’ve done studies comparing training an AI model against a consensus from multiple pathologists with the performance of individual pathologists, they find that AI does a better job of calling mitoses than we do.
MD: And so I think for specific tasks like that, I think AI may end up being helpful where it can either say, here’s a hotspot of mitosis. This is where you should be looking, or it can give us an idea about how many mitoses are on a particular slide. I don’t think that it’s going to replace experts in saying this is the diagnosis without having a human being come back into the loop. And in veterinary medicine in particular, I think that’s for a few reasons. One, we deal with such a wide variety of species that it would be difficult to train an AI to cover all of those.
MD: And even if you restrict it to just dogs and cats, it’s always a little surprising how different the same thing can look depending on exactly how it’s sampled and how you get the slides from that tissue sample. And so there’s a lot of variables that would have to be taken into account. And then finally, the level of validation you would have to do to say, the results that I get, I trust. I’m not sure that it would ever be cost effective in veterinary medicine, at least not for a while.
DH: I appreciate that’s a nuanced and great answer, and I really appreciate having you on the program. Dr. Dark, I really enjoyed talking with you, and I think I learned a lot today.
MD: Thank you. I appreciate you having me on.
DH: Michael Dark is the UF clinical associate professor and Director of the UF Veterinary Diagnostic Labs at the College of Veterinary Medicine. I want to say thank you to the UF Vet School for their help with this program. To all of you for listening today, I hope you enjoyed the program, and I hope you’ll be back with me next time for Animal Airwaves Live.
