05 July 2023
How IoT Track & Trace Solves the Big Cattle Problem
IoT Leaders with Nick Earle, CEO of Eseye and Brad White, veterinarian, Kansas State University professor, and partner at Precision Animal Solutions
05 July 2023
IoT Leaders with Nick Earle, CEO of Eseye and Brad White, veterinarian, Kansas State University professor, and partner at Precision Animal Solutions
When you’re facing a $500 billion problem, you need an urgent solution.
Bovine respiratory disease (BRD) is the biggest cattle problem in the agricultural industry. What makes it so difficult to solve is that it’s based on multiple bacteria and viruses.
But that doesn’t stop Brad White, a veterinarian, Kansas State University professor, and partner at Precision Animal Solutions, from tackling the problem head on.
Brad joins Nick on the IoT Leaders podcast to discuss:
Tune in to hear how Precision Animal Solutions is using IoT solutions to revolutionise farming.
Join us on the IoT Leaders Podcast and share your stories about IoT, digital transformation and innovation with host, Nick Earle.Contact us
You are listening to IoT Leaders, a podcast from Eseye that shares real IoT stories from the field about digital transformation swings and misses, lessons learned and innovation strategies that work. In each episode, you’ll hear our conversations with top digitization leaders on how IoT is changing the world for the better. Let IoT Leaders be your guide to IoT, digital transformation and innovation. Let’s get into the show.
Hi, this is Nick Earle, the CEO of Eseye. And this week we’re talking cattle. We’re going rural. You’re going to be educated about the world of cattle, the number one killer of cattle in the world, which is something called BRD, bovine respiratory disease or cow pneumonia.
As you’re about to hear it is an incredibly big issue, hundreds of millions of dollars just in North America of lost revenue, hundreds of thousands of cattle dying and even more being affected in terms of ongoing sickness.
And the big thing that I got from it is that this is an incredibly difficult problem to diagnose because it’s essentially a medical problem, but trying to diagnose a medical problem in a cow, a head of cattle where they tend to exist in herds of 500 at a time, just think about how complex that is.
So anyway, you’re going to hear a great story here about a startup that came from professors in university who are very close to making a big bite into that problem. They’ve been working with us for two years and this is one of the podcast stories I’ve wanted to get out there for a while. So with that, let’s get going and start the conversation with Brad White of Precision Animal Solutions.
Well Brad, hi and welcome to the IoT Leaders Podcast.
Hi, thank you.
This is an unusual one. It’s our first animal podcast. It’s podcast number 36. And so we’re going to be doing a little bit of explaining on some pretty high level veterinary terms here so we can set the context for this. But before we do that, why don’t you just take the opportunity to introduce yourself because I know that you’re in academia, but at the same time you’re an entrepreneur creating a company. So maybe you can just set the background.
You bet. So I appreciate the opportunity to visit and discuss with you some of the things that we’ve learned. I’m Brad White. I’m a Veterinarian. I work at Kansas State University, but also I am a partner in Precision Animal Solutions, which is a company that’s outside of K State. And we’re going to talk mostly about what I’ve done through Precision Animal Solutions, which is an area where we have looked to monitor animal behavior to help make better decisions.
And I heard that I was going to be interviewing you and that you were in Manhattan, but I think I got the wrong Manhattan, right?
Yeah. So I’m actually in Manhattan, Kansas. So the Little Apple as we call it, is where I am. So we’re right here in the center of the US.
Right in the center of the US. Okay. It’s very, very different from the Manhattan that I’ve been to.
So what we’re going to talk about, and I must admit, we’ve been partnering now for a couple of years between Eseye and Precision Animal Solutions, but when I first heard about you guys and this project came across my radar, people were using the acronym BRD, and then I had to say, “What does that mean?” IT is full of acronyms, so you learn hundreds of them, telecoms even more.
But BRD, these guys are trying to solve the problem of BRD. And I said, “Look, excuse me, just ignorance, I’m the dumb guy in the room. What are we talking about?” So just in case there’s anybody else out there who doesn’t really know what it is, maybe you can just start there, explain BRD.
Great topic. And as far as acronyms, it’s not just IT. So as veterinarians we use a lot of acronyms, too. So it’s okay. It’s all good.
So we’re all good.
We’ll throw some out today. But I think the essence here is BRD actually stands for bovine respiratory disease, so think pneumonia. And one of the things that we see in cattle is that the pulmonary tract or the lungs is one of their more susceptible areas. And pneumonia or BRD, which we’ll call it today, is one of the more common diseases we see, especially in young cattle as they’re going through the growing phase.
Okay. So it’s essentially pneumonia in cattle and it’s one of, if not the biggest killer. Can you quantify that? I mean, I don’t know, how many cattle get sick? And of those, can you give us a swag on how many BRD kills or affects?
Yeah. So let’s talk a little bit, and we may take just a step back just to be sure everybody’s on the same page as far as the cattle industry. So it’s a little bit different, different parts of the world.
In North America and many other regions, well, across the world, we have cows that will produce calves. Those calves are then weaned and typically they’re managed to provide them a diet that will produce the best beef after weaning. So they’re weaned when they’re five to six months of age. Once they get past five to six months of age, BRD or pneumonia is the number one disease challenge that they face. So it’s by far number one.
By far number one.
In the US, we have about 20 to 25 million cattle that go through our feeding system, so as we talk about feed yards or a system to provide them that feed. Of those, if we look at the USDA numbers, about 14% of them will get sick during that period. Not all of it is BRD, but that’s the primary thing. Of the cattle that die during that period, BRD represents about two thirds of those deaths, so something related to respiratory disease. So it’s a pretty important syndrome in the cattle industry. So it’s certainly something we want to figure out.
The interesting thing is we have certainly improved our ability to provide good therapy. So we have better prevention techniques with vaccines. We have better treatment techniques with antimicrobials. However, we haven’t been able to fully decrease the amount of BRD that we face.
And if I just do a little bit of maths, one thing I always like to do on the IoT Leaders Podcast is get into the business use case, and we’ll cover the technology and it’s exciting, but business use case is what’s really exciting, if I just … Well, first of all, it’s not just a killer, is it? I understand that even if you survive it, you can have lower yield. So you have less income, you don’t have as much meat.
But if I just do some back of the envelope math on what you just said, if you assume 20 million head of cattle in North America and 14% get it, that’s 2.8 million. And based on how many of that it kills, that’s a huge amount. And I think a head of cattle is roughly … If I’m a farmer, when I look out at my field of cows and if I think of the money I’m going to get, what’s a cow worth to me? If I lose a cow, how much money am I losing?
Yeah. It depends a lot on the year and the cattle cycle and when they die. So I’m giving you a hedging answer, but as a ballpark it would be pretty easy that right now those are worth between 1,000 and $2,000 per head.
So the death loss is important, but one of the other things, and there’s several issues that play into this, you mentioned it but I’m going to go back and reemphasize it, is of those that die, think of them as the tip of the iceberg because the actual iceberg is we also have those that are affected.
And then the part of the iceberg that’s below the water, which most of us know is about two thirds of the iceberg, the part of the iceberg under the water is those that we don’t ever see as sick, so we can’t treat them and they still have impacts from BRD.
And we know this because if we go look at lungs at the end of their feeding phase and we go look at lungs, we’ll find calves that have severe lung lesions or moderate lung lesions and we never observed them as sick. So we had no way to treat them. Yet, if they have lung lesions, we can go back and look and we can say, “Boy, they didn’t perform as well. They didn’t do as well. And certainly we would’ve liked to treat them but we were never able to see them as sick.”
It’s a huge problem. And the iceberg, I guess that’s a good way of thinking about it, it’s actually much bigger than I guess you can even measure. It’s absolutely huge, especially as you say, in retrospect, when you look at their lungs, a huge number of them had it but you didn’t know. I mean, I’m just trying to do the math. Would I be right in this, just in North America, from an economic point of view, this could be a up to a half a billion dollar problem?
It’s a large problem. I haven’t done the math on what that looks like, but let me give you some numbers for perspective on that iceberg.
So what we’ll see from the research side and putting together several numbers, depending on the group of cattle, I may have a one to one and a half percent death loss. In cattle that are severely ill, I may have up to 15 or 20% of that group that I treat. However, some of our research that has looked at lungs in cases like I just described, a 1 to 2% death loss, 20% we treat, I may end up seeing 50% of the group actually has lung lesions that we didn’t see.
So the problem, and just like I’m going offscreen here, we see this part, the part down here is what we don’t see and that can be a big … that subclinical disease. And it comes into, we’ve got to be able to identify those. So that’s not the only issue that we have, but that is an important issue.
That’s an important one. All right. Huge issue. It’s a global issue. We just quoted the US ones and it’s been around for hundreds of years. It’s not a new disease.
Forever. For as long as we-
It’s been around forever. Okay, so continuing the education. So if it’s been around forever, people have tried to identify it I guess, how can you spot, if you’re looking at a field of cattle? And for those of you, there’s a small percentage watching on YouTube and Brad’s background is a beautiful background with cattle there. But if you’re looking at a field, a herd of cattle, how can you spot it?
And how do we look for it today?
… let me take a brief divergence first. So one other thing I want to say about BRD in general, to give you some background, it is pneumonia. And most of us when we think pneumonia, we’re thinking something like flu or something else relating it to us as people, right?
One difference with BRD that is an important distinction is it’s actually a syndrome. So there are actually multiple bacteria, multiple viruses and multiple other risk factors that play a role. So it’s not just one thing because the logical thought might be, okay, if we have BRD and it’s caused by one pathogen, why don’t we create a vaccine for that pathogen-
There you go. Yeah.
… and fix it? Right. We can’t because there are multiple things. So we do have several vaccines that we use to go through the process to try to prevent it, but they don’t prevent all cases, which means when we have cases we’ve got to identify them.
So that’s back to your question and your question is, what do we do now? And a lot of it is, remember we’re working with animals, cattle that aren’t going to tell us when they’re sick. And in fact through evolution they’re actually attuned to not tell us because in the predator-prey relationship they would be a prey animal.
And they’re a herd animal, which means you don’t want to separate from the herd because there could be something that comes along to see you and identify you as sick, which you don’t want. So they’re attuned to stay together as a group. They’re also attuned to not show signs of illness. We would be considered a prey species. So there are times that actually, as I go out, they may change some of their behaviors.
And the behaviors that we’re looking for are typically, if you said, “Hey, I’m going to go out with you and we’re going to show you how to find sick cattle,” we’re going to be looking for cattle that are depressed. And I’m going to use the term depressed here in the medical sense that they’re not moving around as much, they may have their head down, they may not be feeling good.
They’re anorexic, so they’re not eating. Well, if I’m going to observe that, I actually have to observe for a longer period of time because of course all of us go through periods of day where we’re technically not eating or anorexic. They may have an increased respiration rate, so they may be breathing more rapidly. And then finally if we can get them in a facility, we can handle them, we’re looking for a increased rectal temperature.
So those are some of the basic signs we’re looking for. But as you can imagine, I’m going out, I’m looking at a group of cattle and I may not be able to identify a single one in that group very easily if I have a short period of time to monitor them.
So it’s behavioral. A lot of the symptoms are displayed through behavior changes, but particularly difficult, as you say, well, when they get sick they actually, sometimes they go into the middle of the herd because they’re scared. I mean the predator takes down the sick one at the back, so they go into the middle of the herd. So it makes it even harder to identify.
So you’re looking for something that’s really hard to identify in animals, not people, that’s a number one killer and also affects all the rest of the iceberg. And yet, when you’re looking at them or when they feel concerned, which I guess they would be if you walked into the middle of a field and started looking at them because you’ve got to get close, they then change their behavior and clump together, which makes it even harder to find them.
So you think about the problems that we have in business, these are completely different parameters than we have in business problems.
I was going to ask you, I heard the story that we talked about, it’s been around forever, people have been looking for it, but is it right that there’s a whole industry out there of people, cowboys I guess, who sit on horses and cattle and just try to spot it and then try to separate them and rope them or whatever, that that’s actually a big industry?
Yep. So as we manage cattle, there’s a couple things that we know, and I want to break out a couple things you said that are really important that you touched on. One, we can’t always find them accurately. And in fact when we look at the research, the person is the most accurate tool that we currently have. However, the person is far from perfect. And there’s two mistakes we can make. One, we could not see a sick calf, or two, we think he’s sick, but he’s not really sick with BRD.
And talking about that one first, think about that, the clinical signs that I told you, depressed, not eating, increased respiration, those are not very specific signs. Any type of illness could cause those same signs, which means I could bring him up and maybe I don’t give him exactly the right therapy.
The ones that we don’t see, it’s hard to see because as you mentioned, there’s a group of people that do a really good job and usually they’re checking these cattle once, twice, three times a day to observe for signs of illness, but they’re looking at them in large groups. So they may be looking at a group of 300, 500 head and going through trying to look at individuals.
Now, those that we don’t identify as sick early in the process, if that illness progresses … And think you or I, right? I mean, I may come into work in the early stages of having a cold or having an illness and you may not pick that up.
If three days later, I’m coughing and hacking and lying in the corner, you don’t have to be a doctor to go, “I think something might be wrong with you.” But at that stage, my treatment is going to be a little less effective. So can we identify them early in the process?
The other thing that you said, and I didn’t mention this earlier, but often what we’re looking for is cattle, part of that depression is they don’t stay with the group, they don’t stay with the herd, they’ve eventually gotten to that illness part. But what we found through some of our research is if you actually monitor their behavior daily, hourly, what you find is early in the illness part, they actually spend a lot more time at the herd.
They spend a lot more time trying to hide in the group early in the disease process, which there’s no way for me to tell you as an observer to say, “Okay, ride out into that pen and go find the calf that’s spending more time with his buddies.”
I’m sure listeners like me are reminded in a way of the last three years with COVID and the number of attempts that we had to try and identify people and stop disease spreading.
And we had these, I don’t know about you in the US, I think you had the same as us, in the UK we had these apps. You had to have the app and it was actually saying, have you been within two meters, in our case, so just six feet or so, of somebody else? And then these hugely complicated algorithms are all trying to … Not even. It’s all to do with if you put your own information in that said you tested positive, then who have you met in the last few days?
So the point being is that these are incredibly difficult problems. I mean you are not trying to solve an easy issue here, which kind of explains why no one has really given it a go. So let’s pivot and let’s say, okay, we described a really difficult problem and yet, technology can play a part. And actually we’re at our second POC on this, proof of concept on this project and we’ll talk about that. So we think we’re homing in on solving this.
So what is the basic science here in terms of, maybe to explain to listeners, how you guys started it off? You were at a research company I believe, and you had this idea that there was a way of solving this. So what was the idea or the way of detecting this actually? What was the idea about how technology could play a role in doing this? And then we’ll get onto what did you do after that.
So all of the signs that I’ve described so far are subjective. And so I worked with my partner and we still have a contract research company where we do research looking for better ways to manage lots of diseases, but really a focus on BRD in cattle. And one of the challenges is, if I can’t appropriately identify sick cattle or when they got better, it becomes very hard to show which products work better or not. So we said, “How could we objectify this?”
And we started the process by looking at what are some of the things that we measure. And almost everything I’ve described so far you can imagine is behavioral. So we said, “Let’s use what we can, whatever existing equipment.” And we tested a lot of stuff and we wanted to measure lots of different behaviors of animals. And then we went back and associated those with animals that we knew were truly sick.
And what we started to find was, hey, those sick animals, not only are they behaving differently towards the end of their illness, there’s actually a lead time and we may be able to find them a little bit earlier and more accurately in the process.
So that was the key, as you said earlier. And of course the earlier you get them, the earlier you can treat them and the earlier you can stop the spread like in human health. You used the keyword there, behavioral.
And in this podcast series, IoT Leaders, we’ve had a bunch of healthcare podcasts recently, more and more actually, and the overall trend on all of them is they’re all different use cases, but all of them are trying to move from reactive to proactive to preemptive through continuous monitoring, and you’re describing the animal version of continuous monitoring. But we’re not talking here about wearing a device on your arm or whatever.
When we started, we put devices on their legs, which I would not recommend.
Nobody enjoyed that process, us or them.
So you actually copied the … Yeah, you put devices on their legs and that didn’t work. So what did you do? You’ve got to have something attached to the animal I guess, and you’ve got to be able to identify the animal and there must be a big database and a whole series of algorithms and all the magic is in the algorithms. So maybe you could just lead people through what you did and what you were trying to do and we’ll get onto where you’re up to.
Yeah. So our goal was we started with the problem and we’ve described that problem here and how could we best monitor animals to determine if they’re sick or not? And we tried a bunch of things. So I mentioned we did try things on their legs.
And we didn’t take the approach of, what’s feasible technology today? We took the approach of, how do I collect a bunch of data and figure out which pieces we need? And then after we figure out which pieces we need, then we’ll get that technology and make it work.
So we tried several things including accelerometers, realtime location systems, monitoring even things like cattle temperature, other aspects to try to figure out what pieces of data that we need.
As we went through that process, we started putting together and we started eliminating pieces of data that weren’t helpful. So you could think of in the early stages they were equipped with lots of stuff. We just grabbed lots of stuff and collected lots of data and we said, “Okay, this is not helpful, put it away. This is not helpful. Okay, here’s some of the base of what we need.”
And when we evolved down what was most important to us, were their behaviors. And I’ll lump some of those behaviors into a couple different categories. One, activity, so how much they moved around. So did they move around a lot or a little? Just like you and I when we’re feeling ill, our movement patterns change.
Two, the proximity to feed and water. None of the devices that we used measure if they’re actually eating or drinking, but because we know that feed and water are in a defined area, then certainly we could get feed and water.
So those first two, there are technologies out there, which will let us determine proximity and movement, but the third thing that we found, which was actually very important and we’ve already touched on it a couple times here, is their social behavior.
And their social behavior actually is an important key to the process because how they interact. Because again, they’re a herd animal, instinctively they’re going to have some behaviors that would be different than what you and I do, and those behaviors are going to be modified even more when social pressures are high.
So one of the things that I’ve said is, and if we start piecing this together and think about the background here, I’m going to go back to we said there’s a person that goes out horseback, he’s looking for which cattle are sick. And one of the things that we’re saying to look for is if they’re not eating. So when might you want to go look? At feeding time, right? So that’s when you would go look.
Well, actually at feeding time those social pressures are the highest. Think of the times that you’ve been sitting in a room or a meeting or you’re with a group of people and you may not be hungry, but everybody says we’re going to lunch, you get up and go to lunch, even if you’re not hungry.
So what we’ll see is those social behaviors are important to monitor, but now we’re getting into there are times of day where I can’t see much, which means we need to have something that not only monitors behavior, but continually monitors behavior. It can’t be just a snapshot of the day, we need the whole movie.
And we need to be able to take that data. And that’s one of the things that we learned through the process is even our sick cattle at some times a day appear normal, but if we monitor them the whole day, we’ll actually catch that pattern or difference in their behavior.
A lot of the people who listen to this podcast, and there are a lot of people actually who listen to this podcast, which is great, obviously they have a tech background and we’ve done a lot of explaining on the use case and the agriculture and this whole new world for me and I expect for a lot of other people.
So from a tech point of view, as I mentioned earlier, we’ve been working together for a couple of years as companies, maybe just give an overview of what tech you are now trialing and what stage you’re at in the project to try and get that data that you’ve described so well.
This is I think a really interesting part of the story. So we approached this and I’ve had some others that we visited with that we approached it, what I would say is backwards. So we didn’t start with, here’s the technology, let’s go apply it to cattle and try to answer this question. We started with the problem. We’ve described that, we needed to measure behavior. And then we identified what do we need to collect.
And we tried for a while using off the shelf technologies. We tried using some things that we could purchase and then put on cattle and monitor. We even tried using some things that were made for cattle but didn’t quite suit our use case and needs.
So when we first started our conversation with Eseye, we said, “Okay, how can we get some of these things that we need?” And the cool part of this process and where the relationship has come is the nice thing was the response wasn’t, “Okay, use this, this and this.” It was, “Tell us more about the problem. How can we figure out what we can do that’s custom to your use case?”
Since we had already had the background of what we needed, there was a series of conversations and that relationship has been really important because it’s evolved. So where we’re getting to now is not only … Because what we haven’t mentioned and very frankly, what I underestimated the value of is the environment that we work in is not easy on hardware and equipment. So we have to have things that are suitable for that environment.
And there’s two components there, of course the weather and the durability and all those factors, but also we can’t have wires, electricity strung everywhere because we’re in pretty extensive management environments. So we’ve worked with Eseye to come up with some solutions and technology that meet those needs through an ongoing conversation.
And in fact to shed a little bit more light on that, from our perspective, what we’re doing is, as I mentioned earlier, we always start off with the problem and then we say what the technology needs to be, which is the right way of doing it. It’s hard to do, but it’s the right way of doing it.
So just to bring us up to date, we are as a company, Eseye are designing from scratch, custom devices that are built for this use case. So they are ear tags. You moved from the leg to the ear. I think at one point you were putting stuff around their neck.
Yep. We’ve put stuff almost everywhere you can put stuff.
Everywhere, yes. It sounds like a dangerous job for me with these big cattle. But anyway, we’ve now got custom ear tags that use Bluetooth that has a lot of technology in it. They talk to each other so we can see how close the cattle are.
There’s also receivers on the feeding trough and the water troughs so that we can actually measure how often they eat and how often they drink. We’re talking about the pedometers so we can actually measure like a Fitbit, a few steps. A cow has to do a certain number of steps a day. And all of this then goes to a specialist router or router as you call it in the US that we’ve designed for you that will go on a fence post and the data gets aggregated.
And that’s why you need the universal ubiquitous connectivity because you need every aggregation device to be transmitting and in some pretty remote areas where if you put a proprietary SIM in there, you may find that there’s just no signal. So you need to be able to connect to any network that’s available. So all of those problems have to be taken away. So that’s what we’re doing. So we’re actually creating brand new products to solve this problem.
And you’ve done some trials, haven’t you? I’d love you to share if you can, the results. So I think that one of the key indicators of the trial, I’m going to get onto what you can do with the data, but one of the key indications of the trial was the percentage detection rate. And I think from the first POC that we’ve done where we had fairly clunky equipment, we’re now shrinking it down as I said, but just the first trial, I think I’m right in saying it had some pretty impressive results, right?
Oh, absolutely. And I think as you described it, this is a process. We’re moving through this evolution and because our technology needs were such, there were several hurdles to go through. And one of the things that I have really appreciated is the ongoing conversation. So we’ve had some others in the past that here’s the hardware, make it work if you can. We’re not hardware people. We need that relationship with somebody to have that discussion.
So the first trial that we did, we knew was prototypes and we said, “The form factor is not important in this trial. What’s important is figuring out which pieces of that technology work and can we get the connectivity.” And we found two key takeaways from that trial. One, we had connectivity with the tags. The tags did what they were supposed to do and they worked calf to calf and they worked through the receivers.
Two, from those data, we could come in and we could process and get the numbers we need to make our algorithms run because we’re not the only company that’s looking at how can we monitor and figure out BRD. However, we are novel in the aspect that we’re actually looking at multiple things. There is no singular silver bullet, so to speak, or one gold test that we’re going to use. We have to build the entire picture.
So I described it earlier as the movie and I think there are several things that would give us some snapshots and from those, just like if I showed you a snapshot, you can make a guess at what’s going on, but if I showed you a movie, you have much more information about how things are building and progressing. And that’s what we’re doing with several different technologies.
And your detection rate I understand, you said earlier maybe it was like two thirds in general. After your POC, can you talk about what improvement you saw?
Over the last several years, we’ve done a series of trials. And so we’ve done trials in the US, Canada, Australia, and we have looked at how does the system … And we call it, and I should have mentioned this earlier, we talked about Precision Animal Solutions, but the system itself put together, we call it REDI, R-E-D-I, remote early disease identification.
So that ready system when put with the appropriate technology, we’ve actually tested it and then followed those calves through. So I mentioned earlier that if you follow them through, you can look at their lung lesions at the end and we know the right answer.
What we see is this system actually identifies sick calves more accurately than a person and identifies them earlier than the person would. So on the magnitude of between three to seven days earlier than the person would, which as you can imagine, one of our mainstays for treatments is antimicrobials. So if we have an infection that’s building, the sooner we can treat that infection, the better off we are, which we have indications that’s looking like that’s going to give us better results after our treatment.
So in our previous trials, that’s what we’ve seen, which means when we’re using the antimicrobials, we’re doing it in the right way to treat the sick cattle that need to be treated at the right time when we get the most effect.
Right. And that three to seven days is hugely important in human health. It’s really important from what you’ve said in animal health because of the problems of diagnosing. And like I said, the cost because unlike humans, when an animal dies, you can actually put a number on it. And we swagged it, at 20 million and 1%, it costs about 2,000 for a head of cattle, it’s at least a 400 million, within other lost income and whatever, it’s probably a half a billion dollar problem in North America alone. And then three to seven more days or earlier diagnosis, you can see the business outcome ROI on a project like this and that’s North America.
And we focused on BRD, but there’s other diseases that cattle get. There are other behavioral disease. There are other things that we might be able to pick up by behavior. Now we haven’t dove into those yet because we got to get the technology right and we want to work on the biggest thing first. But I think the potential is well beyond just looking at respiratory disease.
And I do want to make the point, I don’t see it supplanting some of our current people that are doing those jobs. I do see it augmenting their ability to do their job better.
And probably make the job nicer, frankly, because now you’re being told go find number 526 as opposed to stare at 800 head and-
… try and find one. There it is. And by the way, it’s called 526. So I would’ve thought the job would be a lot nicer for the people.
It’s interesting what you just said. We did a great podcast with the CTO of Biofourmis who are doing continuous monitoring and early diagnosis of heart disease. And Milan Shah, one of the things he said at the end of it is, “Well, it’s for heart disease, but what we’re really doing is we’re building up a database of human behavior. It’s anomaly detection, machine learning, AI, cloud platform. And we take all these,” they call them biomarkers. They take eight measurements and they combine them together into groupings and they create a hundred biomarkers.
And he said, “We’re finding that we can absolutely diagnose early onset heart disease much earlier.” But he said that actually when we give the data to people who specialize in other diseases, they say, “I’ve waited for … I’ve never had data like this.” It’s physical symptoms of disease. And okay, you’re the heart guys, but they want to look at diabetes, they want to look at a whole bunch of other stuff.
And I guess that’s what you’re saying is that once technically you can solve this problem of getting the data into the cloud for continuous monitoring, then the opportunities to use AI, ML, the opportunities to analyze that data, anomaly detection behavior is really broad. I mean it goes across the entire spectrum. So it’s bigger than just BRD, this opportunity.
It is, and this is where I think the value comes into play is that it takes multiple components to capture that value. So A, we have to have the technology to capture that data. B, we have to have the ability to do the machine learning and some of the other aspects.
And then the final corner of the triangle is, we have to understand the applicability, which of those pieces of information can I use to make decisions? How can I impact the outcome if I knew the answer? But you can’t have any one of those components without the others, which is why our partnership with Eseye has been valuable.
Well, it’s very nice of you to say that. And I said at the beginning, I wanted to do this for a couple of reasons. One is it’s an unusual one. In our world this is an unusual one. I think it’s a great example. It’s a huge problem. It’s a massive ROI. It affects cost of food, it affects animal welfare. It affects everything.
But the other reason, and you were kind enough to mention it earlier, as we finish here, is that we have this unusual approach of people … We’re the crazy people in IoT. People say, “Really, I’ve heard about you. I want to talk to you, Eseye.” And they say, “How much is your SIMs and data? Show me as usage. Show me your current products and how much they cost.”
And we’re the crazy people who say, “No, I’m actually not going to do that. That’s not what we do. I want to ask you about your use case. I want to understand your problem. And then what I’m going to do is I’m going to propose a solution for you. And it might be that we have to design it and build it for you, but I want to solve your problem, not sell you what’s on the truck today.” And you said how much that differentiated.
I think in healthcare that is particularly true because if we look at all the healthcare ones that we’ve done, well, we get involved in the design of the solution for every one of our customers, but healthcare in particular now, the biggest thing happening, because we get to see a lot of patents at our level, is the movement from proactive, preemptive. There’s a supply chain or there’s a process chain and by the time you’ve got to identify and try and sort it out, it’s often too late and you get these terrible percentages that you mentioned.
Continuous monitoring is going to revolutionize healthcare for people and for animals, but the way in which you have to gather that continuous monitoring data requires new hardware, new solutions. And it means that you do have a two- or three-year project of design before you actually get to deployment, which a lot of people will say, “Oh no, I’ve got …” You can get to deployment next month. Here, I’m going to ship you some products that’s off the shelf. Like you said, there’s a ton of ear tags already available.
But ultimately the success rate and the way to solve it properly is to actually start by designing the solution for the use case. And it’s really great that after a few years of doing that, we’re now getting these podcasts. And I really thank you for coming on and sharing and for your partnership because I think we’re in year three of a relationship, we probably got another year before we’re really out there deploying in anger, but based on the results so far, this is going to be really big news for the cattle industry and for farming of cattle.
And I should also do a shout-out. It’d be remiss not to do a shout-out for our partner, TELUS, the Canadian operator because I know TELUS Agriculture were involved with you as well. And they came to us and said, “Can you build a solution?” So it’s been a partnership with a very visionary operator in the form of TELUS.
But a year from now, Brad, I mean based on what I know of the project timescales and the trials that you’ll be doing in about a year’s time, you could be really making a significant contribution to food supply chain, the cost of food, animal welfare, yields of cattle. I mean it’s a really nice, ambitious, but nice project and it must be pretty cool to be working on, especially as you came from academia and you said, “I’ve got these theories, I want to prove it.” You’re not far away.
Yep, that’s right. And we’ve got a good team. I mean, you mentioned the team. So we’ve worked with … We’ve got some partners at TELUS. I’ve got some other partners that are veterinarians and it is not a single person can … As you mentioned, it’s ambitious and our team is really great about putting some of these things together and being willing to try new things.
Well, maybe we’ll have you back on the podcast in about a year or so when-
That’d be great.
… you can actually say, “Hey, we’re there. It’s working and here’s the real data.” You could become our first repeat visitor.
Listen, we’ve been going around about 45 minutes, which is what we normally do. I really wanted to thank you and thanks for everything that we’re doing. Good news is the project is on track internally. I’ll give you that update. I’ll share it with everyone. As I say, it’s the first really animal one that we’ve had, which just shows the potential of IoT is so broad.
I think people would’ve really appreciated not only listening to this but learning about your industry and the problem that you’re trying to solve, which makes some of the problems that we deal with on a day-to-day basis look easy. At least I can talk to people and get communication back in some of the problems I’m trying to solve.
So Brad, thank you. Thanks for being on the IoT Leaders Podcast. And thank you to our listeners. I hope you enjoyed this episode and we’ll be back soon with another one. But in the meantime, thanks again to Brad White of Precision Animal Solutions.
You’ve been listening to IoT Leaders featuring digitization leadership on the front lines of IoT. Our vision for this podcast is to be your guide to IoT and digital disruption, helping you to plot the right route to success. We hope today’s lessons, stories, strategies, and insights have changed your vision of IoT. Let us know how we’re doing by subscribing, rating, reviewing and recommending us. Thanks for listening. Until next time.
We hope today’s lessons, stories, strategies, and insights have changed your vision of IoT. Let us know how we’re doing by subscribing, rating, reviewing, and recommending us. Thanks for listening. Until next time.
Build the IoT estate that meets your needs now – and ten years from now. It’s why global leaders trust Eseye.