Deepak Sahu 0:06
Thank you. It's good to be here is the last we had a lot of data discussions in the presenting companies. And to discuss about this, no better than the people who are here we have Caitlin Morse says the CEO of BrainSpace. Having revolutionizing the ICP market, and the cerebral fluid market, simplifying the treatment and still collecting a lot of data to change the neural space. We have Dave Chiriac is partners, Windham ventures also has his previous role as a provider. So he comes up with an angle, does he really need the data that we are talking about? We have Dave, Dave Cuban Keo Capital Advisors, create mind, create, invested into health tech, has invested in the digital surgery platform, case index, doing well in the European market will bring us the European concept of the data. And Peter, one is CEO of neutral mix. When we talk, we always talk about how do we do continuous diagnostics, monitoring. And he brings in with a lot of passion and collecting a lot of data with his own platform today. So before we go in there, we just I just did a small scan about the presenting companies that presented this year. And the mountain of data that we are collecting as an innovators within this LSI conference. That is every device today collect some form of data. So so the first first topic that Caitlin when we were discussing earlier brought in is worth the adaptation of data looks like we have such mountain of data, what will make this data feel adaptable being used by all the stakeholders in the healthcare system?
Caitlin Morse 2:19
Yeah, I think it's a really interesting question that has evolved with the last 10 years, right? Because we saw this, okay, we can produce data, we can get a continuous stream of data. And even a lot of health systems early on would say, well, we're gonna build a data lake, we don't know what we're gonna do with the data yet. But we want to make sure we own it, we want to make sure we have a copy of it. And so we're gonna put it over there. And at some point, their AWS Bill started getting large enough that they started going, Hey, hold on, maybe this isn't the right way to approach it. And I think if we can really actually look back towards the Industrial Revolution, if you want to look at the the mining of raw materials versus the finished goods that come from those raw materials, right. And so, as an industry, we really started and saying, Hey, we have steel. And now what we really need to say is, what are the machined parts that are needed? And what are they going to build? Right? And so when we look at data, the question is not just can we produce data, but what value are we going to create from that data. And so I think the med tech industry really has an important role in that process. So if you think about even the evolution, you know, all of us have become a lot more familiar with large language models. And what that looked like 10 years ago, people were saying, we can use natural language processing on nursing notes. And we can try and figure out what's going on. And anyone who studied languages knows that's nearly not so easy to do, there isn't the syntax that exists in other languages, and we have this concept of devices, then humans are recording that data. And then another device is trying to read it. And then software is trying to decipher it. And really, we're moving into this now newer era and med tech where we're able to say, this device can build the data, put it in a structure, make meaning of it and communicate with other devices. And humans can spend that time face to face with humans. And I think a perfect example of that with what Peter's working on with nootropics is to say, rather than somebody placing an order and sending it to a lab and it coming back, and maybe this maybe that right, is to really say how do we take what is a continuous stream and remove from that the pieces that are most relevant?
Deepak Sahu 4:11
We do good segue to you. How do you think about your data is gonna get adopted?
Peter Vranes 4:19
Yeah, I think it mean data for data's sake is useless, is a backward step. So we have to be very considered, if we're gonna get the trouble of building a device that's going to generate data. It has to be it has to change in action that a clinician is going to take that's going to improve an outcome, because that's what we're all about, improve the outcomes. If we're not very crystal clear on that. Why do it doesn't make sense. So we have a technology to medical wearable, and we can measure any diagnostic target continuously and in real time. So that gives us Lots of options. And we have to get really good at thinking through. What's number one? What's number two? What's number three? And so what are the criteria upon which we make those decisions? And the data? Is, is how we make that and we think about what what don't the clinicians have now? What gap? Do we fill with this data? How does it impact their decision making? And how does that improve an outcome? And that needs to go, we have to think through that for all the applications. Otherwise, you have data for data's sake, which helps no one. So it's a very considered process from the start. This is right at the start, before we ever build anything we think, we think through that process. And then we go and produce the evidence that says, well, let's validate and quantify those outcomes with that data. So that's, that's the hierarchy of how we think through this. And, and we apply that to, to lots of lots of different problems we think we can solve with this tech. Amazing,
Deepak Sahu 6:14
but we saw the passion of two innovators out here. So I will ask both Dave, you, and Dave, to you to to put on the caps of an investor. And think about what actually makes sense. When do you think about investing in a data driven med tech platform?
David Kereiakes 6:34
Yeah, I think that's the benefit of having two days as I can divide. Yeah, so I have been investing in in healthcare for 1314 years, mostly in devices to start of my career and saw the rise of a need to digitize healthcare, knew I had to get smarter in it and appreciate what all the acronyms mean, because they get very defensive, and challenged in a very defensive, yeah, so I moved to Providence ventures, where I helped manage the health systems venture investment arm as well as the Digital Innovation Group, which was 150, software engineers, software developers, data scientist, all employed by providence, the third largest nonprofit health system, to get a better sense of how software really influences and can be used in health care. It was ran by the gentleman who launched Kindle for Amazon, and a former Amazon executive. And then half of the management team or C suite at Providence is former big tech. So a very novel, strategy and practice when the CEO of Providence started pulling all this together in the city of disruption in Seattle. And it was novel to bring a device investor to sit next to and with the digital innovation group. And it really shows just how siloed things are and how fragmented health systems are with devices with data with software. And when you start talking about that convergence that's happening and needs to happen with smarter and more intelligent devices that communicate can help inform clinical pathways help, lower cost, drive, greater outcomes help qualify the patient make these very challenging clinical decisions. Devices are incredibly well positioned to be able to enable that and do that. But it's, it's challenged by having to sell through supply chain and through the administrative leadership through the leadership of the AOR, on top of a clinician that it just wants that device and to be able to use it. And so one of the biggest challenges that you have is the user is often different than the purchaser, and your ROI and the value prop has to adjust based on who you're talking to. And so, again, data for data's sake and being able to just pull something out of a procedure doesn't always mean something. So I was thinking of it took me back to business school where a large burger chain spent a billion dollars and this was a long time ago. In a software platform that you you're like what we're talking about healthcare, right burger chain, but that knew and could tell, centralize the exact temperature of every stove, every grill that they had around the world into one central location, their headquarters to make sure that all the girls are at the exact temperature. It meant nothing. Like they did it. They built it all the temperatures were there. Quality was the exact same standardization and two degrees, three degrees made no difference whatsoever. And they immediately scrapped the entire system. So and then when you look at, there's plenty of other these case studies where you're you're trying to find something valuable in the data, it is far better, and far easier to define an ROI, what it is that you're actually trying to pull out and get to inform that decision. To make it easier for a healthcare leader, depending on whatever department they're in, that has been in that seat for probably 20 or 30 years. It doesn't cross into other areas of the health system to get educated enough and understand and comfortable enough with why you're trying to sell them something. So
Caitlin Morse 10:51
what if the data wasn't part of the ROI? So I think we have an opportunity in devices, where if the device itself is treating is given is diagnosing right is doing something already, there's an existing reimbursement code, there's an existing workflow, and the data that's being collected, you're only showing the value of later, does that something you've seen, I mean, NHS being a classic example where they're not going to pay just to collect data for data's sake, right?
David Cubbin 11:18
Maybe it did help to take a bit of a step back, first of all, so yeah, we all agree that there's a wealth of data out there, we need to figure out what the you know, what the outcomes and effects the the analysis arm, we invest into health tech, were a family office backed investment firm. And, you know, we focus quite a lot on surgery. And so, you know, in spite of all the data out there, in spite of all the fantastic technology, surgery is still the second most dangerous place in the world, outside the battlefield. So depending on how you cut it, we're talking about a 0.5% risk of death, which equates to the, you know, the most dangerous year in Afghanistan, it's 16,000 times more dangerous to go into surgery than to catch a plane. So the risk of complications is up to 15%. A lot of this can be fixed with data. The problem is partly who's collecting the data, and to what aim or to what effect. We believe in the power of the platform. Not the burger chain platform, necessarily, but we believe in the power of the platform to you know not to take data from surgeons and tell them, you need to do this, or you need to do that. Which you know, can be self serving at times. But we believe in empowering hospitals and empowering clinicians to make decisions in their best interests and in the best interests of of, of their patients. That's
Deepak Sahu 13:17
completely right. What we heard over here. We also did a small discussion with a lot of surgery departments across the US. And the clinicians always told us data is the tech is not the barrier, the trust is the barrier. Right? So what we all talked about is insights and evidence, and to David's earlier point of ROI. And what Caitlin also raised, what if there is no ROI? And that brings us to the next topic of thinking about how should we monetize this? Who is gonna pay for it? So we have a mountain of data, we generated insights. And there is a very useful case around it. We were inside ICU, there was a device that will give us alarms to the nurses. Too many alarms lead to nurse fitting. So how much is good? Who is going to pay for it? So I'll start with David tune with you. Because you are in the NHS system, you come from a different system. Casein Tex is generating revenue. Yeah, what made made the NHS pay for care CINDEX.
David Cubbin 14:39
The NHS has not paid for syntax. But I'm very happy to say that because I mean, I couldn't be happier to say they had fun. But actually, a lot of a lot of systems countries, you know, hospitals, insurance companies have paid for the product. So the The reality is that you can get to break even you can get to profitability, once you've reached scale. And, you know, the US is, is a different beast, every hospital system is different, and healthcare systems different. And so we are in hospitals in the US as well. But the reason why we've got to where we are, is because we have been able to demonstrate that ROI. And that's the most important thing. It's, you know, this is all outcomes based. So, in the case of insurance companies, I mean, the operating room is still a complete black box to two insurance companies. And so yeah, and we're talking about trust. And trust is such an important issue as it relates to data. big issues with trust and insurance companies, people are paying more, and they're getting less. So you know, how do they change the game, it's through understanding the risks, and being able to provide, you know, more more dynamic pricing and more specific pricing based on understanding their patients better, but also understanding the risks in the operating room better. So we're well supported by the insurance industry. And in the case of hospitals, yeah, of course, hospital margins are under a lot of pressure globally. It's, we attend these conferences, and there's, there's all this amazing technology, but you know, who's going to pay for it, so the costs are going up to pay for all these wonderful things. So if I look into the future, and we'll talk about the future later, how do we get there? Well, I mean, we know that all this stuff has to happen, you know, the data has to be collected, the system has to become more efficient. There's not a Choice. But as we improve hospitals margins, through measuring the right data, cleaning it up, organizing it, analyzing it, we can drive up revenues, we can improve margins, we can reduce waste, can do all these fabulous things. And and people buy and if you don't want to buy a product, fine, we'll share the cost saving, you know, that's, that's, you know, the the proof is in the pudding. So, yeah, yeah,
David Kereiakes 17:32
so who pays for it? I mean, there there is a potential storm on the horizon for innovation. And for small companies like what we what I have built a career off of, and what hopefully will allow my kids go to college if they do and are able to, is how it well, I, in a provider, I was approached by large med tech that wanted to, and we're at a point right now where payment models are changing, and they're getting more creative, because the the business of delivering care has a negative margin. And it's if we continue to deliver care in the way that we have, hospitals will go out of business, and just the delivering care out of the towers that we have. And so a lot of these larger device companies are looking at taking risk and bundling all of their devices, and giving an ala carte to the provider and saying you have one throat to choke, and supply chains are getting incented to consolidate vendors. And so as a new novel technology, it's harder to get in and differentiate yourself when you can't sell against free, because the large medtech will say they may be better, they won't say that. But you may be beating them head to head in a particular device, but they fill out the rest of that procedure. And they will say we'll help qualify you on the front end on what patients should be eligible for this particular procedure will follow them after with data and and insights. And by being able to predict who's going to respond or who whether they have diabetes, or their health score, how it whether they smoke, and if they have a third floor walk up, do they have a ride home? Those kinds of unique insights can help inform how a patient is going to actually do in a procedure. And they'll go at risk. And so for our health system that's struggling with margins, and with all these vendors in supply chain to say, Great, we'll write your one check. You take care of that. Our clinicians will be happy because they'll be able to pull whatever bag they want out. And I mean, maybe not because they they want something new and novel to be able to deliver but there's there's this potential storm out there. And I think if that happens, and if we aren't innovating and showing how we can position novel technologies to would own a particular ROI and drive, the reason why you should be paid or why you're taking costs or enabling a procedure to be done in a lower cost setting or out of the towers. It's just something to think about and and be aware of. There's,
Deepak Sahu 20:15
there's a really good point Caitlin when we were discussing earlier, we discussed about devices that use data, and devices that generate data, you are two very different business models that you are thinking about many data companies that come into the market, think about that they're gonna make money based on the data the very next day, right. Talk to us a bit about brain space. And what are you thinking about doing with data and your launch sequence?
Caitlin Morse 20:47
Yeah, so I think David does bring up a really important point. And if you're a startup founder, who doesn't know how big companies sell into hospital systems, find someone who will tell you the stories because knowing that that's an approach they take, knowing that they'll often go yeah, we'll give you a certain, you know, the capital equipment for free. If you do a certain type of disposables, or knowing some of the ways that they go about these tactics, you need to be able to play at that level. And so that often means having, for example, positive unit economics pretty early on, so that if you're giving away material, or you're providing a 90 day study, you're doing whatever, that's not digging you too far in the hole, you need to be able to make a clear value prop, that's not just head to head against those guys. So it is really important, if you aren't already familiar to understand who's in your space and the kind of games they will play to maintain the position they're in. I think one of the things that's really been important from our perspective is I've seen a lot of places, if you are in a major city, you have access to the best clinicians and the most academic centers and the cutting edge technology. If you're not, and I'm not, I mean, I've spent time in in very resource constrained environments, and some of those are right here in California. If you're in an environment where that's not available, you're often not getting the cutting edge technology. And it goes beyond that, you know, what there's been a lot of data around, women being excluded from clinical trials, or different ethnicities not be included. But there is also a divide between academic centers and the rest of the country. And so one of the things I think there's really an opportunity to do if we're going to digitize this is to actually drive clinical studies happening in places that do not have the resources to do it. If they had to do it all manually, we're able to have research being performed that there is just no clinical ROI for so I think getting creative about what is that data producing is part of it. In our particular case, BrainSpace is a device that both generates data and uses that data in therapy and diagnostics. So we're not just collecting data. And we're not just monitoring we're also making, we're also responding in a closed loop fashion. So doctors and nurses are interacting with our system and saying drain to a pressure target drain to a volume target behave this way or that way. Let me know about these different conditions. But in the meantime, we're also automating workflows that are currently nurses, we're already looking at how we can be freeing up ICU beds, right. So really looking at what is the clinical workflow? What is the hospital choke point, and a lot of people will tell you that if they can't get people from the ER into the ICU, they've got a problem. If they can't run surgeries, they've got a problem, right? So really understanding that broader context of if hospitals have negative margins, what are you doing to help that? And then similarly, if you're able to collect that data passively, then you know, we were saying earlier, you can't do data for data's sake. Well, you can if there's no marginal cost to do so right. And so maybe some of these places where we can build these datasets, there may not be a use immediately, but it can support that research, if it's not causing that additional cost. But of course, if it's going to add to the health system costs, you're not committed to it.
David Cubbin 23:49
Can I just jump in with something very quickly? You I think when you're collecting the data, you do need to you need to be very careful about how you're collect collecting it upfront, and how you're organizing it upfront. Because, and I'll I'll give you a simple example. We looked at a at a robotic business technology that was doing hernia operations. And sorry, actually, I don't want to talk about that one. I want to I want to change the subject. The secret? Yeah, that one's a secret next year, but no, there's a we were looking at the provision of drugs and pain medication around surgery and and we were finding that, you know, people were giving scores of their pain in order to get medication and, and we noticed that some sort of outliers and you know, it was like 123, and then occasionally a seven, and then they'd get a certain medication and what we realized and we we had to drill into the data, we realized that and that everyone was lying, that the nurses were deliberately lying or telling the patient to lie in order to justify the provision of the pain medication. So it is I mean, it's a simple analogy, but it's, it can be a lot more complex than it appears. And
Caitlin Morse 25:18
that's why I think it's an important distinction when it's device generated data, sure, because you are removing the human from the loop, right. So if you know if Peters got a particular value, that is the value, right, what humans choose to do with it then gets into exactly what you're talking about the psychology, but we have an opportunity where the device is actually generating what is at least consistent, right, you can eliminate interoperate availability, you can eliminate some of these behaviors. Now, of course, you still have to decide if you agree with the validity of it. And that comes back to your earlier point about trust. But we do have this opportunity to at least standardize some of what's being collected.
Peter Vranes 25:51
And I just wanted to jump in focus. Just one point you made Carolina about geographical disadvantages that exists country and city, basically, because I think it's a, it's a, it's been a problem for ever, really, when you think about it. And there's been a big trend to remote patient monitoring. So pushing the hospital into the home, it's advantageous for the patient, they want to, it's advantageous for the hospital, it's cheaper, everyone seems to like it. But there's a problem. And the problem is enabling technologies to allow it to actually happen. And so what we're seeing more of are those enabling technologies that are mobile, wearable, perhaps, that allow for people patients to be treated in the home. And when you can do that you decentralize healthcare. And it's exactly what we need to do. Because expecting people to live five hours away, to come in every day into the hospital to get a blood draw to get a lab diagnostic done, doesn't work doesn't work very well, and it's disadvantageous for them. And there's, and there's better ways that we can treat those patients. And
Caitlin Morse 27:05
in some cases, it's going to be actually at home. But in some cases, it might be the local clinic or the local community hospital. Right. One of the things we've been looking at with digitizing some of this is then you can have that expert level one that's at the center of excellence with print within Providence, for example, can go and provide that console, if it's all analog, and it's not available, and it's not clear or like you were saying it's not well structured, and it's not easily accessible, then they're not able to ride that care. And so changing that sit that care setting, whether it's getting them all the way to the home, or whether it's just getting them to a clinician that's a lot closer is is really a huge potential that we have with this type of technology. It's
Deepak Sahu 27:41
a good point, because we do not have to go far if you go to shutter health today. Last year, they appointed someone as a chief medical officer for hospital at Home program. And so the providers are also thinking about pushing those systems. And we discussed about monitors and a lot about from a hospital perspective. Peter, your device could bring those systems to home to could bring to alternate care settings, too. So how do people think about monetizing those data? Because it's a continuum of care. It's not care only at the hospital?
Peter Vranes 28:17
Yeah. Let me give you an example. So say kidney transplant patients. They need immunosuppressants. They've got to be monitoring monitored for creatinine the kidney function biomarker. And they need to do that regularly. And they need to do that for a long time. Now, if you're not close to a hospital, a lot of those patients literally move. So imagine you've gone through a traumatic period in your life, you just had a organ transplant. And now you get to move and you get a move for quite a while and you're going to be close to a hospital with the device that we have the wearable Bluetooth connects to the cloud, go anywhere in the world. So you can have a the doctor sitting in anywhere in the world. And in real time that will see their level of immunosuppressant because they need to get that within a certain therapeutic range for individualized for their patient, they will see their creatinine levels in real time. They won't need to get off the couch to go into the hospital or clinic to get a blood draw. Because we're monitoring them continuously and in real time. And you can apply that imagine you can start to think now. What are all the applications that you can apply that to. Another one is therapeutic drug monitoring of antibiotics like vancomycin. That's a big problem, very narrow therapeutic window, high toxicity. And it's a bit of a guessing game. The way that it's those are the first two days. There is no monitoring. And then they take one blood draw at a trough so And then use that to try and predict what the next dose should be. The outcomes are very poor, a lot of people get toxic doses, a lot of people die from acute kidney injury, it's really cleared. A lot of people get sub therapeutic quantities of the of the vancomycin, and what it's used to treat things like sepsis, MRSA, these are life threatening conditions. So imagine, you know, half the doses not hitting the therapeutic range, because they get limited data, and they get it delayed, it's always in the revered rearview mirror, you take the blood, draw it, send it to a lab, and then hours later, at best, you get a result. Now, where's the patient? When you get that result? No idea. You got one data point? Is it trending up? Down flight? You have no idea. So what do you need to do take another blood draw. And the whole process starts again. And that in reality just doesn't happen? Because they don't have the resources. So so there is a there is a big need for to fix that problem we're talking about the topic is data. Why do we need data? That's that's a good use case for data, right? Well, we're going to fill in the gaps, we're going to do it in real time, we're going to do it could be remotely could be in the ICU could be in the general ward. And we have a continuum of care. Because the same patch that you were in the ICU is the same patch that you were when you go home. So that will continue. So that's just an example a couple of examples of the utility of of data. There's
Deepak Sahu 31:27
a good point because what we are discussing over here is not only evidence, right, we are discussing about health, economic and outcome research as a whole. You talked about various movements of the person that's a time and motion movement of a person from a different different topic. Dave, Caitlin, you wanted to say?
David Kereiakes 31:47
Yeah, just real quick, and I won't distract us too much. But you by Bluetooth, enabling you, we're talking about some clinical now in economic outcomes to sell and give the information to the purchaser, or the payer or the provider on why they should pay for your technology or or what value you can bring. But you also have the ability to be more intelligent about your business that way too. And the data that you can pull in on utilization, who's using it, when are they using it? How are they using it to help your sales force to help your team coordinate and better understand the procedure that you can always be there in, I think to a an investment that I was fortunate to make in a company called Varian medical. It was a lung navigation system to better biopsy and then eventually treat for lung cancer. It was very labor intensive, we struggled with gross margins up to a certain point because the rep needed to be in the room. We solve this data, we saw the utilization trends, and they had to wait and schedule for a rep to be there. And so we created a virtual rep 10 years ago to a clinical rep to chime in on the capital to be able to communicate with the with the provider. And that unique insight building up to it far increased our gross margin and helped with utilization. Doc's pulmonologist, rather than interventional radiologists were doing percutaneous needle biopsies which they didn't and they weren't entirely comfortable with because it can create a pneumothorax, which is an overnight stay and a big complication. So having that and removing that barrier was something that we used in the insights of what we were seeing. And so the data doesn't have to just drive your value prop to the provider can or payer, whoever it can help enable your consumer, the patient better understand how they can utilize your technology.
Caitlin Morse 33:51
Well, and along those same lines, other members of the ecosystem, right, so Peter Deepak was asking about the monetization part of it. And I know there have been a lot of remote patient monitoring technologies that have come out that are sometimes fighting over the same reimbursement dollars, right. Have you looked at all you need to talk about the drug effects? Is this a potential? We're actually it's the pharma partner who wants to get that and maybe interested in some of that monetization or supporting clinical trials? Or are there some creative ways to be able to monetize that that may not necessarily just be the typical reimbursement?
Peter Vranes 34:21
Great question. Thanks for bringing that one up. That's right. Yes. Oh, good. That's it? The answer is yes. Because why? So let's say you're a drug company developing a new drug. We all know that's a very expensive process that don't all get through the regulatory pathway. And one of the reasons it's expensive is because you have big cohorts, the large number of participants going through trials. And what do they have to do? They've got to go into regular blood draws, they're measuring the drug level, they're measuring safety markers. Well, what's a way that we can get and radically reduce those costs, while we can develop a sensor that can measure the drug that they're interested in, and we can develop a sensor, which we have the measures, the safety markers, and then the participants, the 1000s of them can wear it at home, and they don't have to go in and get their blood drawn and get one data point, they can get 1000s and 1000s and 1000s of data points, and they can do an H cheaper. So it's, it's, there's more data, they've got a better chance of getting through the regulatory process, they can show efficacy better, they can show safety better, and it's cheaper. There's a few good reasons why they might be interested there. So these are the sorts of things now that that requires a technology that can be produced quite cheaply. Because we're not going to give them a million of these things, they're probably going to want in the hundreds or maybe a small quantity. And we're fortunate that DNA by sensing, which is what we have, that's our tech, we can produce this exceptionally fast. So and that's where big data can come in as well. It helps us to develop our senses very, very quickly. That
Deepak Sahu 36:14
brings us to the last thing before we open up to the audience for some questions. Are we do we need to go from a mountain of data to an ocean of data? Or do we need to come down just in a single line? What's the future?
David Cubbin 36:32
What I've observed from an investment perspective is that, and especially in the age of AI, you know that this this boom, that we're experiencing is that, in some ways, you don't need as much data, in reality, as people think. But sometimes it takes you 10 to 15 years to figure out what data you need and how to analyze it and what the insights are. So just from the investment standpoint, you know, we we've observed that, you know, the the most interesting companies, not the ones that are saying, AI and this and that, but actually the ones that have been doing this for a really long time.
Caitlin Morse 37:24
And I would argue that both mountains and oceans are the wrong type of analogy, what we actually need are cities, we need streets, we need skyscrapers, we need traffic lights, and roundabouts, we need structure to the data. And when you look at structured labeled annotated integrated datasets, then AI actually has a much simpler job to do, when you're trying to decipher through what's there and make sense of it. That is a much more sophisticated algorithm you better build than if you can say, go to this particular coordinate, and look at what you're seeing and make sense of it in the context of the rest of the data. So being able to provide that structure is something that as device companies, the more that we can create those datasets in a way that they're AI ready, whatever that looks like in your context, the more valuable they can be.
David Cubbin 38:08
And just to add very quickly to that, because you've inspired me is just that a lot of the way data is analyzed today is surprisingly unsophisticated within healthcare for an industry, which is actually incredibly technical and sophisticated. When we're looking at averages. And like there's dirty data, even when it's organized. It's not particularly useful. And then you apply some averages to it. Doctors give up the data and they're saying, you know, what, what am I getting? What am I getting back for this? It's it's really key that we apply the same rigorous data science to the industry that's already being applied in other areas like sport, for example.
Peter Vranes 38:57
I'll be quick because I can say it's less than a minute. Quality number one. sighs number two in that order.
David Kereiakes 39:05
Perfect. Yeah, I would just say this is an incredibly exciting time to be in healthcare. The lights are just being turned on in a room that has been dark. You referenced the disparity in outcomes and the lack of data that we have, if you just look at 18% of participants in cardiovascular studies are women Women make up only 18%. We know how white old white men respond to disease. And it wasn't until recently, we started to figure out a third of women present with different symptoms when having a heart attack, and they're twice as likely to be diagnosed with a behavioral health disorder when they're actually having a heart attack. So this unique insight that we're finally starting to get is the lights are turning on and it's exciting just within the last 10 years. I think it was 2016 McKinsey did a study the digitization of industries, healthcare was right below government, but right ahead of Yes, and you're behind the government. If you've been to a DMV or DMV, and you've seen the efficiencies there, they've gotten better. But in that time, it was right ahead of hunting and agriculture. And I guarantee you agriculture jumped above that very quickly. So it's exciting to have 20% of the GDP $4 trillion global spend just in the US. It's the fourth largest global economies, US healthcare spend, to finally start seeing this light in a room and shining lightness is exciting. Sorry, I went over No, it's
Deepak Sahu 40:35
perfect to end this. When we were working with the mental health organizers in mental health digital app, the CEO told me right, is it if I can have a pricing study for my Medicaid population, I can make it available for all the peers. So it's not the data is what you do with it. But just anyone in the audience any question for for the team?
Guest Question 41:07
Hi, my name is Jessica. I'm in ophthalmology. And I guess one thing I just want to add is multimodal data, like incorporating different aspects of imaging and also, you know, lab values, etc. But something of course, everyone is going to talk about as well as all this data cybersecurity, and how are we going to protect ourselves against maybe in the future, there may be a cyber war and that's kind of where warfare is going.
Deepak Sahu 41:36
So, being coming myself from a cybersecurity background, the number of breaches in the healthcare systems are humongous. So, we are already under everyday in under an attack. Okay. So somewhere somewhere some healthcare professional lives the the Drive Open somewhere, some it's also happening from human mistakes, right? So the cybersecurity has to evolve with the evolvement of the of whatever we want to do. To add on to that mix, we have aI coming in, we have other data's data sets coming in, and there is not an enough infrastructure with the computing power we have. Where do we need to cool the policy has to involve the policy of sharing data, the policy of HIPAA, and all those has to evolve, if we have to tackle the cybersecurity point.
Caitlin Morse 42:37
And just go to absolutely agree, I think that's part of why structured data is so important, right? If you can have timestamping, if you can, I mean something that basic right, this is sometimes multimodal data is simply the fact that you're at different time intervals. So really thinking about what are the other pieces of data that a given clinician is going to be looking at? And how do we interplay between that, the more that that can be done at the device level, rather than having to create a software layer on top to reconstruct it, the more efficient that computing power is going to be. That was
Peter Vranes 43:06
what add one thing, the most important priority of all of this, in my view is not actionable insights, although that's critically important, it's trust. The day we lose that, that the it's it's it's game over. Because if people do not trust us, whoever the provider, the hospital, to be good shepherds of their data. That's when we really got to have issues. And these data breaches that we have over and over again, each one of them is an erosion of trust. And I think we don't do half of what we should do in that area. Because once trust is lost, it's very difficult to get back.
There was a sobering note. Sorry, I just I felt I felt that was a bit heavy.
Caitlin Morse 44:06
I have a very hopeful question.
David Cubbin 44:08
But the war hasn't been lost and the future
Deepak Sahu 44:15
so any further questions, because we're gonna go and have a wine after this anyways, so we're standing between them and half er, a little harder to get another question. Thank you. Thank you.
© 2024 Life Science Intelligence, Inc., All Rights Reserved.
