All right. Let's get this started. The pitch competition. One of my favorite events of any conference is always exciting to see the companies lined up. And as usual, there are some great companies out there to start off with. My name is Kadir. I'm with JGDC, the venture capital arm of J&J. And I'm involved in investing. It's a pleasure to be here and invite all the great companies up there for the pitch. So before I call the first company on, my request to you, please stick to your time. There's a timer out there in front. Ten minutes for the pitch. And then five minutes for our eminent judges to ask some questions here. So with that, I'm going to start off with the first company. iCardio is not here. So we're going to start off with CareLog. Hello. Can you guys hear me? Yes. So it's an honor and privilege to be here. I'm Aman, the cofounder of CareLog. And we are developing ECG-based foundation model. And we'd love to tell you more about the work that we have done over the past year. So I would like to start by highlighting that heart diseases cost a lot of money and one of the biggest problems today. We see two important parts of the problem. The first part is that patients show up symptoms quite late and the disease has progressed quite a bit. The second part is the diagnostic workflow that we have for these diseases is quite staged. And depending upon the symptoms, we use greater tests. So but what if we were able to have the diagnostic ability of a CT scan, of a BNP blood test, of an echocardiogram, potentially on a smartwatch? And that's the future that we are trying to build towards. The big problem in achieving this vision is that you would need massive amounts of structured labeled ECG data and CT scans done for the same patient on the same day. And this is the big blocker in terms of building out these algorithms. So what we are trying to do is we have developed the ECG based foundation model with the goal of reducing the amount of labeled data that it would require to develop these algorithms. So today we have a 200 million parameter ECG model and what we have been able to do is achieve algorithms with just 10% of the volume of the data. We have designed it in a modular way so that you can input any number of leads that you want. Obviously these are leads, a subselection of leads from the 12 lead. We have not tried it yet on other devices, but potentially we can move it. And when we compare the architecture to other approaches used in the industry, at least in our internal set, we have found better results. So this is sort of how we develop the foundation model. The central idea is that you need to have a pre-training task for ECG signals, which is specific to ECG, and the central idea is that you mask a portion of the ECG and ask an algorithm to complete the ECG signal, and the idea is that you can get to some features and after you do this pre-training task, then when you want to fine tune an algorithm, you would require much less data. So on the bottom I have shown, so the task that we have is low ejection fraction detection from an ECG signal. So on the bottom you would see that with 5,000 pairs, we were able to achieve a similar AUROC as compared to 50,000 pairs. So if you see the comparison of architectures, so with our foundation model, we were able to have decent accuracy, even at 10% of the volume of the data. We have also tried to do this with just 500 ECG and echo pairs, and still have had decent results. On the bottom left, you would see the AUROC for the low ejection fraction detection algorithm, and so this was the first task that we took, but today our portfolio has eight algorithms which have been derived from patients that have had an ECG and an echo done on the same day. So this is a data set of 50,000 patients collected at two hospital systems, and we have eight parameters today, so low ejection fraction, left ventricular diastolic volume, left ventricular systolic volume, atrial enlargement, we also have BNP elevation, we have right ventricular dysfunction, and ventricular hypertrophy. So these are all parameters that we were able to find out from ECG and echo using our foundation model, but what we want to do is, we want to pair it, we want to create an ecosystem where we pair different ground truths with ECG signal, and just find out, like, is there discriminative ability on the ECG for other indications. So because we have the foundation model, and the volume of data that is required to develop these algorithms is quite less, so that makes it accessible for smaller centers in different countries in the world to be able to partner with us. So we are trying to create this ecosystem to get on different hospitals on board, to get different innovators on board, and just fine-tune algorithms with, let's say, 500 ECGs, 500 labels, and make it really accessible for different people to develop algorithms with our technology that we have developed. So we have some brilliant clinicians working with us, trying to study different aspects, so there's a lot to unpack here in terms of the technical approach and also the fine-tuned algorithms and how to fit these algorithms in the workflow today. We are a very young team, basically four engineers and a physician, and we have been working hard since last year on this project. Dr. Jack Singh is our clinical advisor, and we have Dr. Charu Ramanathan, who has been a commercial advisor. In terms of the development that we have had up till now, so we have finished the clinical validation of low EF on 16,000 patients. We are looking to apply for a 510K later this quarter. And we have signed a commercial agreement with a company called Bunker Hill Health, which have successfully commercialized a CAC screening algorithm out of CT scans. So what we want to do is we want to build a business model around incidental diagnosis, as in partner up with hospital systems, run these algorithms on routine ECGs, and find additional patients for the hospital systems. And that's the business model that we want to build out. These are some sample numbers for if we were to have a product which was focused on low EF, but we still don't have proper clinical data on the financial ROI yet. These are just estimates from references that we have seen in research. So apart from the eight algorithms, yes, we are looking to expand our portfolio. We have applied for a breakthrough designation already for a diastolic dysfunction algorithm. We'll be applying for a 510K in the coming quarter. And what we want to do is raise a seed round after we get our first FD approval, build an ecosystem, expand the amount of indications or amount of parameters that we can get out of an ECG. And we'd love for any of you to come back and partner with us and discuss further. Thank you. Great. Aman, thank you. That was Cardiologue. I appreciate that. Thank you for keeping us on time. I just want to remind the audience that you can submit your questions on the HRX app, and we might be able to come look at that. Maybe before I turn it over to the judges, it wasn't very clear for me on what the unmet need that you're trying to go after. I mean, I understand the various analysis that you're trying to do, but now what are you trying to replace or what is not working very well out there that you're trying to replace? So the vision that we want to build is we want to have the diagnostic ability of stronger imaging tests that are normally done inside the hospital on a relatively simple signal, so an ECG signal. So today we are focusing on patients or parameters that are derived from an echo. And we are able to, for example, detect low ejection fraction from an ECG signal. So we want to increase the accessibility of these imaging tests onto a simpler test. So the idea is if the community had the diagnostic ability of a CT scan on a watch, what implications would that have? So we are tackling it in a very broad sense of manner. There are multiple interesting parameters that we have come up with. For example, diastolic dysfunction. Diagnosing diastolic dysfunction just off an ECG, that means you can only diagnose people before they get into heart failure with preserved ejection fraction. And hopefully, just with the pharma treatment, they will be able to have better outcomes. So that's the overall vision that we are trying to go after. Thank you. Excellent presentation. One question and comment. As a cardiac imager, I would say to diagnose diastolic dysfunction by echo, we end up with a third of cases that are non-diagnostic based on ASC criteria. What echo criteria are you using to feed into your algorithm, number one? And number two, for BNP cutoffs, as you know, those vary based on BMI. And the consensus is that if your BMI is over 40, you should be reducing that BNP cutoff by 40%. And are you doing that with your algorithm? Yeah. Thank you for that question. So for diastolic dysfunction, basically, we took the echo guidelines and derived the ground truth as per the echo guidelines. So I think we have four parameters, mainly it's the E2A ratio, the TRV, the size of the atria indexed with the body surface area, and another parameter. And we use the exact diagnostic algorithm that a clinician would use on an echo. And we use that to derive our label for diastolic dysfunction. And then finally, we are taking ECG as an input, this ground truth derived from this echo criteria, and training an algorithm that way. For BNP, the algorithm that basically, if you have discriminative ability on the echo, I think you don't even need a BNP. But we basically just ran it to show that there's discriminative ability. Because if you think about it, there's a central disease state, which is showing up in an echo, which is showing up on an ECG. What we want to show is that what you are traditionally used to seeing on an echo or a BNP test, it's possible with algorithms to see on an ECG also. So that's where we want to actually get to. So two quick questions. Number one, if you take a step back, what are you doing that is uniquely different from others who are trying to do the exact same thing in the field using EKG signal, ECG signals to predict? And secondly, as you think of how to change the care pathway, I want to go back to the first question that was asked. What has to be true for the signals you're measuring to actually reduce or change the use of echocardiography? So on the first question, basically most of our work has been focused on building a foundation model specific to EKG. So today, most of the technical approaches are taken from images or transformer-based approaches which were actually natively developed for language modeling. We have been focused on developing an approach which is native to ECG as a signal. So that has been our major area of focus. And the reason for that is if we are able to crack this successfully, so today we have a model with 200 million parameters, we have shown some data on that we are reducing the volume requirement. That makes it extremely accessible for different clinicians to partner up with us. So for example, if you go to a small hospital in India, they would not have 1 million ECG and echopairs, but they would have 500 ECG pairs with other diagnostic tests. So that's what we want to enable. So we want to enable an ecosystem so that this technology development is fairly accessible. And it also opens the possibility of a variety of use cases around finding treatment targets. For example, maybe finding a biomarker on the ECG with just 500 patients on if a patient would respond to a particular treatment or not. On to the second question, today what I see is immediate impact that we can have with our parameters today is on the front of resource utilization. So we have great NPV. So a lot of these echoes done at the hospitals return normal. If you can already say with an ECG that, okay, these patients probably don't need an echo, are normal on these 10 parameters, then that's a use case that can be opened up. I think all of these use cases have to be studied in detail, and we are really on the first step of the journey to be doing that. Could I just ask one comment? Understanding that this is an early stage company, but one of the things that would be really helpful is for you to not only go beyond the competition, who you're competing against, but also the fear is that you don't want to be a company with a technology that is looking for a home. You want to see if you're understanding the problem that you're trying to solve, and you're the best technology to actually address that problem. And I didn't get a clear sense, because one of the things that you haven't developed yet, and maybe you just couldn't present it, is the business plan to get there, and defining who the customers are, and who wants to use that technology. And because there are a lot of, again, technologies and solutions looking for a home, and you don't want to be that. By the time that you get to a point that you're trying to commercialize, you may not have the resources to do it. So you've got to think about that earlier on, so you could start to really focus. Last comment I'll make before we have to move on is related to what David and Rob said. It would be good to really make a compelling case as to why you are a product and not a feature. AI is often a feature. Obviously you have product potential, but I think having a strong argument as to why you are an independent product that can be used in multiple verticals with a strong revenue pathway I think would be compelling. Thanks a lot for the feedback. Great. We have one question, but I think we're going to keep moving, and instead of time. So audience, if you have any questions, then the presenters will be here after the presentation. So please do stop by and chat. So with that, I'm going to move on to the next one on the list here. HyloMorph. Good morning, everyone. My name is Simone Bottan. I'm the CEO and co-founder of HyloMorph. Our goal is to stop infection in cardiac rhythm management patients and beyond, and we want to do that by delivering on your table the right tools and innovation to prevent infection. Sure, you're all very familiar with these pictures, but I would like for today that we look at it from a different angle. This is state-of-the-art 21st century technology to avoid infection. That hasn't been a breakthrough in decades. Basically, all we can do is to apply the aseptic control measurements and deliver antibiotics, antibiotic prophylactic systemically to our patients. And despite all the best effort we can take, we can still not guarantee that when patients go home, they will be fully out of danger. At HyloMorph, we believe we can do better and that we can advance infection prevention by doing two main things, by delivering antimicrobial drugs right on top of the implants locally and in a sustained way instead of systemically, and by setting the stage for the use of newer, more potent drugs to fight antimicrobial resistance. The EP community has pioneered advanced infection prevention. Believe me when I say that there is no other community out there that has come even close to achieve what has been done in here. They should all be looking at you as a reference for the scientific rigor and robustness that you have put in understanding infection. Today we know the exact incidence, risk ratification, and economic impact of infection in our patients. We know from large clinical trials, randomized clinical trials, and real-world evidence data that between 1 and 5.6 percent of our patients will develop an infection depending on risk. This makes it up to 1 in 18 high-risk patients getting infected. We know that this impacts mortality. Infection can spread to the heart, get systemic, and kill the patients. We know that the patient would spend days and weeks away from work and family because of that. And the treatment costs are huge as a compound can go up to $150,000 per patient. You have put together clear guidelines from the international communities on how to treat and prevent infection. And that's excellent. However, we also know about reports showing that not always guidelines are followed. And in a recent study from the Medicare patients, we know that 4 in 5 patients are not directed to the proper guidelines. So there is still room to do better. From an industry standpoint, Medtronic is the only player out there that has done something meaningful in advancing infection prevention. Their product, Tyrex, is an antibiotic diluting envelope that dilutes rifampin and minocycline around cardiogridon management devices. It has proven clinical and cost benefit from large clinical trial, 20,000 patients in total, summing up the RAPID study and the PADI study, have shown this. The adoption has grown exponentially in the last few years, thanks to this data, and the reimbursement are becoming available in various health care systems and geographies. However, Tirex is mainly used only in high-risk patients, and while I do acknowledge that Medtronic has done an extraordinary job in bringing it this far, I also recognize that Tirex is, they have failed in progressing any farther, as Tirex is still the same as it used to be 10 years ago, and the antibiotics it uses are 50 to 60 years old. At Idemorph, we have created Vesta, the world's first fully resolvable, sorry, the world's first modular, fully resolvable antibiotic eluting envelope to deliver drugs right on top of the implants locally and in a sustained way, so right where it is needed at the time when it's needed. With Vesta, we can not only match the state-of-the-art in antibiotic elution, but we can unlock the potential for the future of this field, mainly thanks to three key differentiators. Number one, its modularity and flexible design allows the loading and dosage of virtually any drug. Number two, the soft tailoring or form-fitting factors. It feels very different from what we use with Tirex, and it opens the door to using many other implants, and last but not least, it is very cost-effective to produce. In Vesta, we combine patented technology that we have adapted for various industrial fields. We make them our own from the polymer manufacturing industry, textile industry, pharmaceutical formulation, industrial design. We have adapted this technology, made them medical-grade, and combined them in Vesta. Today, when we use Vesta around an implant, this stabilizes the implant in the surgical pocket, meaning that it minimizes migration and mobilization. It elutes antibiotics for seven to ten days above the minimum inhibitory concentration, and then it is fully resorbed after about three months when the healing has happened and the functionality has been achieved. It all seems very easy when you see it on a slide, but trust me, it requires a massive amount of know-how to combine all these things and make it work, and especially to keep costs under control. I have prototypes here with me. I'd be happy to show them and demo them to those of you who are interested. Our journey started a few years back. We are actually a spin-off of the Swiss Federal Institute of Technology, where Michael Fanders and I met, doing research on biomaterials to improve the biocompatibility of implantable medical devices. With our first product, we have gone from lab to first-in-human in less than five years, and there have been a number of pivotal moments in our journey, among them the Innovation Award by the European Arrhythmia Association that happened at the same time as the VRAPID and PADI trials were becoming available. We've had also several interactions with the FDA, a full submission done last year, a clear feedback obtained both by the agency as the large industry player that we have interacted with, so by now we have all the information needed to move on. We know what needs to be done, and how to get to the next big milestones, and today we have outlined a clear path forward to achieve that. We are currently fundraising to achieve FDA clearance and U.S. market launch in the next two years. I don't shy away from saying that we are one of the very few teams out there worldwide that possesses the full technical, regulatory, clinical, and entrepreneurial know-how to navigate the complexity of infection prevention. It's a very complex field. We need to have both vertical and transversal know-how. We're working with more than 60 people from external providers, mainly in Europe and the U.S. We put together a strong network of active collaborators. We have experience in similar fields and track record. Some of them are also on our board, and last but not least, of course, we work with clinical advisors who deal with infection on a daily basis. But we don't want to stop here. Our goal is to impact infection prevention worldwide. We see the category of management devices as just the tip of the iceberg. We foresee a not-so-far future where all industry manufacturers will have to provide advanced strategies to prevent infections in their implants, so we have already made plans to move, to bring vessels in the neurostimulator space, and we know we have the potential to move in several other indications from there, including orthopedics, reconstructive surgery, ventricular assist device, and so on. We also have a number of pipeline projects, including the use of AI for infection control and the use of another antimicrobial drugs, but time is short. I cannot dig into those. In closing, I want to send a message, ladies and gentlemen, as part of the cardiac implant management community, we, Hyalomorph, and you doctors, nurses, investors, industry, and regulators, are sitting together on the driving seat to make changes in this field, and as such, we together have the privilege and the responsibility to lead the way, also for the other communities. I want to leave by quoting Dr. Bruce Wilcoff, who was so important for us to get where we are today, and who passed away this year. Our patients need CID therapy to live. What we need to do is to make those procedures as safe as possible, so we need to drive a change in mentality, and we have to accompany that by the right technology, so if you share this, our passion, and the shared view of the world, please get in touch. We're not only looking for investments, but also for early adopters who want to embrace this mission and go on with us. Thank you very much. Great, thank you, and again, thanks for being within time. There's idle mark. One quick question before I turn it over to the judges. So, the trials for anti-infection, as you know, at least to get to claims, can be long and expensive. So, based on your, you know, what your platform potential, do you think you might need to, from a regulatory strategy point of view, new trials for each device, each indication, or do you think there is going to be one trial that where you can get paper submissions for subsequent applications? So, it's on two levels. From a regulatory standpoint, we can apply the 510K regulatory principle to use the same drugs as used in Tarex, and the same indication as used in Tarex to get regulatory approval, but the specificity of the carrier rhythm management, the data of Tarex, of the rapid trial, are sufficient to demonstrate the benefit of the technology. The moment you want to move into newer drugs or newer indication, there would need to be, yes, work done specifically for indications, and nevertheless, you can do that on two levels. Number one is we have to disclaim that there is a difference in preventing contamination of bacteria and then preventing infection. Infection is a more complex and manifests in different ways. So, the first level of showing and demonstrating that you can prevent contamination, as much it's a lower burden to achieve, while the prevention of infection, yes, requires a large clinical trial. But the good thing, it is time-consuming, yes, but today we know exactly what protocols and how to, what approach to be used in there. So, the effort that has been done in the rapid trial from by Tarex is helpful and can be translated, I believe, in the newer indications. You, it was a nice presentation. You, I have a couple questions. You mentioned you do it for less money, right? So, what are your unit economics? What's your cost? What do you think your gross margin is going to be? And what are your sort of company aspirations? So, having a single product and developing a sales team in a highly fragmented market is challenging. Obviously, that's why Tarex makes sense to be part of a larger device company. But now, if you go to a device company, you may limit, be limited to a single kind of device or part of the body. So, then would you use a distributor? I mean, it'd be interesting to understand kind of your growth aspirations and also the unit economics as to why you think you're more competitive. All right, yeah, so the unit of economics, we start as a reference selling price as comparable to Tarex. So, we're talking about about $1,000 per unit as reference. We think that's a reasonable price, especially for high-risk patients. What we're working on is to reducing the cost of good as much as possible. We think we can achieve and be twice as more cost-effective than what Tarex is today, thanks to the manufacturing process that we have. And about three times more cost-effective than the other products, collagen products out there. About company aspirations, we are a mission and purpose-driven company. So, our goal is to achieve as many patients as fast as possible. And it is obvious that to get to this goal, the best way is to team up and partner up with industry manufacturers. So, yes, the ideal scenario for us is to an asset transfer, a licensing or an agreement with a large corporation to get into as many patients as possible. It doesn't have necessarily to be, you know, for all the technology. We think we see clearly that manufacturers of CRM devices are not manufacturers of breast implants, for example. And then you can sort of move and that's the aspiration here. So, David, before you get your question here, I want to make sure we address this and other presenters, if you're not planning to address, make sure you do. How much are you trying to raise and will you commercialize in EU and then US strategy? And what is the ETO sterilization or sterilization methods? Yeah. So, we're currently looking at a financing round of about 10 million US dollars to achieve the FDA clearance and an agreement with one of these corporations, as I said. We're also doing work for the regulatory clearance in Europe. That takes a little bit longer because of the medical virus regulation in Europe requires pre-market clinical trials. So, that has to be added on top. So, approximately there are two years shifts between FDA clearance and European clearance. And regarding this sterilization method, that's the other question, we use irradiation to sterilize. David? You said 510k? Yeah. And I'm a little surprised because usually implantable, especially one that is in the body for a while, you're going to be able to demonstrate, you know, the various different effects on the body. So, I'm surprised that it's a 510k. Or you use, is it a de novo or is it a 510k with predicate? I'm just surprised by that. Yeah. Don't be surprised. It's not that uncommon, even for implantable devices. The moment you have a predicate device, and in this case we have with Tarex, it can be done. It is discussed with the FDA and agreed in pre-subs. Yeah. And yeah, so... And just following on Kadir's question, is that, you know, there are situations where when you use different drugs, often the FDA want a drug-device combination type of discussions, or perhaps even separate paths for each of the different drugs. Did you face that with your pre-subs? Yeah, not directly in the pre-subs, but there has been an ongoing conversation that I've also been having with the FDA folks these days, because it's obviously a topic. Yes, they involve people from multiple expert centers. So we have the CDRH for the device, and you see, I forgot the name, and the people from the drug for evaluating the drug. The moment you include a new drug, so an existing drug with a proven benefit, to be confirmed with the FDA, but you can still maintain the same type of indication to prevent bacteria colonization, and to demonstrate that your drugs are effective in killing bacteria. Then the moment you want to move into an infection prevention type of claim, that's a different story. But the moment you want to bring a new drug in the market, that's also a different story. Thank you. We're gonna move on to the next one here. Sorry, keep moving. Cardio-diagnostics. Are you able to hear me? Yes. All right, so I'm really excited today to be here. I wanted to thank everyone for inviting us, and to be a part of this, and this group of innovators. And I guess, as a cardiologist of the past 30 years, one of my greatest frustrations has been the large number of children who attend my appointments with their worried parents, when the child has absolutely nothing wrong with them. And this happens very, very frequently in pediatrics, perhaps less so in the adult world. But the problem that bothers us the most actually causes anxiety for about 1.7 million parents every year, and costs Americans about five billion dollars annually. And what is that problem? Well, in Alex Trebek's words, it would be, what is a heart murmur? A murmur is just the sound of blood flow, but it comes in two flavors. It comes in innocent and pathological. And for innocent murmurs, there's actually no structural heart disease whatsoever. The patient is totally well, and doesn't require any medical attention. The pathological murmurs, they actually have structural heart disease that need medical attention. For the pediatric cardiologist, seems like a simple problem. You use your stethoscope, you use your ears and your brain, and you have a listen. You go, what are the acoustic signature of this murmur? Sure, this sounds innocent. Nothing else is done. You send the patient out of your office. Unfortunately, that happens nine times for every one patient that you see that might actually have cardiac disease. So why is it a problem? Well, it's a problem in primary care. Primary care physicians are great at identifying the murmur. Their sensitivity is not bad. The problem is their specificity and their positive predictive value. That's really low. So in the end, you end up creating anxiety because they can't explain to their patients what the murmur is caused by. And they have to be able to say that it might be heart disease, which then creates anxiety, which then pushes the the parents and the physician to want to seek further medical attention. So by doing that, you have this huge conundrum where you have a large population of patients with murmurs, 80% of them, 80% of children have murmurs. Only 2% have heart disease. So you end up with these large numbers of innocent murmurs referred for investigation and assessment. And in numbers, we're talking about 3.2 million people, children in North America having murmurs every year. New patients. That results in 1.7 million being referred to pediatric cardiology and cardiology for assessment, echocardiography, when there's nothing wrong with them. All of this to potentially pick up the 61,500 roughly patients who actually have heart disease in North America each year. So what do we do? What we did was we came up with a process and a product that places the pediatric cardiologist basically in every primary care physician's hands so that they no longer have this doubt and they no longer have to create anxiety in their patient. What we did was we patented an algorithm that fundamentally deconstructs the clinical skills of listening to a murmur by a cardiologist. We broke down what murmurs are into the fundamental physics and the acoustics of innocent murmurs because fundamentally that's the problem. You need to be able to recognize an innocent murmur. So what do we do? We took the signals, pass them through this algorithm, which replicates the cochlea of the cardiologist. We then fed that feature into an AI and an AI platform that then allows us to, with great accuracy, pick up and identify innocent murmurs and thereby identifying all the pathologics because if they're not innocent they're going to be by definition pathological. Just what a cardiologist does. And the platform's really simple. You use a digital stethoscope. We're device agnostic so we're not tied to any specific device. You pass it through our application in an iPhone. It goes to our cloud where our AI platform then mimics the skill of the cardiologist, listens to the sounds, and then determines whether it is in fact a pathological or an innocent murmur. There should have been a little slide there that showed what this app looked like but apparently that didn't show up. Well it did. Anyway, it's too late now. So what is the goal of our algorithm? Well the goal of the algorithm, AI algorithms are common. So what's special about ours? What's special about ours is it's built from the ground up based on understanding the physics and the acoustics of innocent murmurs and how the cardiologist actually discerns these sounds. And we add that to the machine learning process, the pipeline. And by doing that what we could show was that we had a 35% improvement in the specificity of the detection of pathological murmurs which makes it actually as good as a clinical cardiologist, pediatric clinical cardiologist. And many times better than your primary care physician and certainly better than just a traditional machine learning model using neural networks. So that's the value of the patent. What does it mean? It means that we reduce the number of referrals potentially each year by 1.4 million. That's the equivalent of three billion dollars worth of care. And we still refer 290,000 patients and we still pick up 10,000 more pathologies and decrease the number of misdiagnosed by the same. So what are the key benefits? Well, the benefits here are societal. We have a huge impact on how health care delivery happens to the pediatric population and will reduce anxiety amongst our population, reduce the time and cost of care in our health care systems, but we still improve detection and reduce misdiagnosis, which is fundamentally the problem for the primary care physician. They are making these referrals because they have doubt, so we remove that doubt for them. This also allows you to have accessible care in places where specialists are not available, and so this is again a very important feature of having this kind of technology. So how do we make this work? Well, right now we think this is a societal issue. This is a fundamental issue in the health care system. To try and deliver this just to primary care providers is working from the bottom up. That might take you a very long time, so we propose that instead we launch this as a health care initiative. We approach governments, we approach health care providers, so insurance companies, HMOs, Medicare, federal agencies that have an interest in ensuring that there's an efficient and effective care delivered at the least cost, and then we provide a cost to our technology of about $150 a case, which is far less than a consultation with a cardiologist and certainly far less than doing an unnecessary echocardiogram. This then would give us an addressable market of about $480 million annually for the 3.2 million murmurs that come out every year, and if you include follow-ups for monitoring for murmurs, that market could be as much as $1.5 billion, and that's just North America. So globally you're looking at potentially $20 billion. Most people say that in pediatrics there's no market, and that's the problem. So why bother with it? Well, it's not true. It's in fact how you address the problems, and what problems are you addressing? You got to find the problems with the most impact and address those. We're really proud of the progress we've made. We're a young company. We started in 2021, and since 2021 we filed a patent. We've collected the data that is where these numbers come from. We started a trial in multiple cities in Canada, and we have connections in the U.S. and in Saudi Arabia where we're going to start collecting data as well, and we're hoping to put that information and data and publish that, and then get that towards our FDA approval. So the team basically includes me, Robert Chen, and my partners Shamir Iqbal, who's our AI expert and has had experience in developing startups, and then Santok Dhillon, who's our chief medical officer, and he's an experienced electrophysiologist, and then of course we couldn't have done any of this in such a short time without partnerships, and so we've had partnerships with our hospitals, the IWK Children's Hospital, which is in Halifax, Nova Scotia, on the northeast coast of Canada, and especially proud of our lobster, and then of course our government agencies. We are affiliated with the provincial government, who has helped us through their health care, Nova Scotia Health, and Invest Nova Scotia, which is also helping us with a safe round at the moment, and then Sparrow Bioacoustics, who's really key to us because they're going to be helping us potentially have direct-to-market. Anyway, I'll now open up for questions, and thanks. Are you raising? How much money are you raising? So far in non-diluting capital, we've actually done all this one on next to no capital. We are great at... No, how much are you raising? Oh, how much are we raising? We're hoping to raise about a million to one and a half million. Okay, all right, the judges. Excellent presentation. Question, are you using Echo as your gold standard when you are considering doing a clinical trial using this, not just the ears of the pediatric cardiologist, or is that the comparator? Yeah, so when we go for FDA, what we will be doing is, we won't be a novel device. We'll have a comparator, so we will be using devices like, actually, Litman, and in nowadays Echo as well because they already have approvals as well. The difference is that, you know, for us, we are actually seeking approvals in the pediatric community, so it's a little bit different, but yes, we do have actually a comparator, so we do, we don't have to start a clinical trial as such, but we will have to do clinical studies, absolutely. Yeah. All right, is that better? Yep. So, first up, really nicely done, a nice description of the problem you're trying to solve and how you are uniquely doing it. One thing about the math, it went by quickly, but it almost looked like your opportunity was bigger than the amount of money you were going to save, and so one thing to think through as you think about the workflow, as you're trying to kind of reduce the use of unnecessary echocardiography, is how does this workflow work, and so how do you prevent every single visit to a pediatrician becoming $150 as they listen with a stethoscope, because they're not going to distinguish the use of this, they're not going to necessarily wait till a murmur occurs, or, you know, and so, does the math get to a point where you're actually costing the system more because of how you're doing this? No. So, I would just work through that, because at first blush, my reaction was your opportunity is bigger than the amount of money you're trying to save. Yeah, no, it actually doesn't work out that way, because when you have this number of murmurs, murmurs are detected incidentally, and so if the problem is the incidental detection, every time a kid goes into an emergency department, they hear a murmur, and they send a referral to cardiology, right? Child shows up for a fever, they hear a murmur, gets sent to cardiology, so this is where that 1.7 million comes from, and so if you look at that math, well, no matter how you work it, at $150 of cost, you're going to inevitably save roughly two to three billion dollars annually, because your cost initially was five billion dollars to look after this population. I just have a simple question, what stops a very simple approach of recorded audio being sent to a cardiologist for a yes-no to completely disintermediate your model? Because you can't scale that. You don't need to, okay, so if you have a health system where you have your friendly cardiologists in the ER, and they can record acoustic, and be like, and they send them a text message, hey, can you listen to this, is this good? Yeah, how many cardiologists do you need? Because the problem here is in pediatric cardiology, for example, in North, at least in Canada, we have a thousand pediatric cardiologists for 40 million people. In the United States, pediatric cardiologists, I think you have a thousand pediatric cardiologists for a population of 350 million people. You're never going to have enough pediatric cardiologists to look after innocent murmurs for the cost, and still have them look after congenital heart disease. No, my point is, is that the unit economics may not scale at 150 if there are other credible threats. That's the thing, my comment would be to work through that in the way Rob mentioned. Okay, quick question from the audience here, if you can make it quick. Answer, how would you address pediatrician concerns about liability in case of under diagnosis? Yeah, so at the moment, you know, our sensitivities is very high. It's 93 percent, and if you think about it right now, at the current system, you're already missing roughly a third of pathology. So you're missing about 18,000. We're going to drop that by about 10,000 a year. So no matter how you look at it, yes, it's not perfect, but then nothing is. And so the question becomes, well, how perfect do you want it? And right now, this is as good as we can make it, but sure, can we make it better? Absolutely. More data, more patience, more time. We will make this better. In fact, for us, this is the first step on a long journey. For us, this is the entry into AI for auscultation. We already have plans to look at, for example, second sound splitting, third heart sounds, and all of this. Things that all cardiologists do in their routine day-to-day activities to identify heart disease in childhood. Okay, thank you. Cardiodiagnostics. Let's have the next person come up. Third Coast Dynamics. Good morning, everyone. Thank you for being here today. We're Third Coast Dynamics, and we're leveraging the power of deep learning to advance aortic precision medicine. More than 10 million people in the United States are at risk for thoracic aortic disease, including more than 6 million patients born with bicuspid aortic valve. More than half of patients born with bicuspid aortic valve will experience progressive dilation of their aorta, which is the largest artery in the body. This puts them at risk for life-threatening events, such as aortic dissection, which is a tearing of the aorta. There is a need to improve how clinicians identify who is at risk for these life-threatening events. The current standard of care is to perform CT imaging to measure aortic diameter size. Shown here are two real patient cases at Northwestern, Daniel and Martin. Daniel and Martin have aortic diameters of 4.8 centimeters, which places them in an intermediate risk category. Yet despite their identical risk profiles, they experience dramatically different outcomes. Daniel remains stable, while Martin experiences an aortic dissection. This is not uncommon. It has become clear that diameters are an insufficient predictor of risk. If Martin's aorta had measured greater than 5 centimeters, guidelines would clearly recommend surgery. Yet at 4.8 centimeters, or the intermediate risk category that we refer to as the gray zone, there is more uncertainty on how to proceed. Patients in the gray zone typically undergo annual CT imaging to measure aortic diameter size. And with each new scan, they must decide whether to proceed with surgery or to continue with medical management and risk dissection or rupture. At Northwestern, 70% of CTs ordered for aortic surveillance are for patients that fall within this gray zone. CT remains the standard of care, despite only providing us with aortic diameters, because it is widely available, low cost, and quick. At academic medical centers, there exists a highly specialized technique called 40-flow MRI, which allows us to visualize blood flow through the aorta. Over the past several years, promising evidence has shown that aortic hemodynamics can help identify patients at high risk. However, 40-flow MRI is not a scalable technique clinically, because it requires a high level of expertise, high costs, and long scan and post-processing times. But what if we could use AI to derive 40-flow output directly from CT? TCDFlow uses AI to depict aortic hemodynamics directly from routine imaging. It was developed by pioneers in 40-flow MRI as a scalable solution to make aortic hemodynamics widely accessible. Our cloud-based platform negates the need for traditional 40-flow MRI technique. It requires no extra scan time, equipment, nor training. TCDFlow seamlessly integrates into clinical workflows. Once a patient's scans are acquired, they are de-identified and sent to our cloud-based platform for analysis. They are then returned to the local hospital network. Clinicians are provided with an interactive aortic rendering, as well as a report with a panel of hemodynamic biomarkers and a risk score. Going back to our two patients, Daniel and Martin, when we applied TCDFlow retrospectively, you can see a clear difference in hemodynamic profiles. Martin has significantly elevated hemodynamic biomarkers that are strongly associated with aortic risk in the literature. If TCDFlow had existed, Martin would be reclassified as high risk and identified as an ideal candidate for surgery. We could have intervened years earlier and prevented his outcome. The ability to reclassify patient risk has many opportunities for creating value. Patients identified as high risk can undergo more targeted prevention and earlier consideration of elective aortic repair. This can prevent downstream emergent surgeries with high health system costs and poor patient outcomes. For patients identified as low risk, we may be able to safely do less. For example, reducing the frequency of imaging surveillance, safely delaying elective surgery, and possibly reducing unnecessary surgeries, which themselves carry operative risks. We have conducted an initial feasibility study at Northwestern in more than 1,700 patients, which showed excellent agreement between our TCDFlow technology and 4DFlow MRI. The next step in product development is to show that AI-generated hemodynamics using TCDFlow can improve risk prediction for patients with thoracic aortic disease. We have received $300,000 in non-dilutive funding, and we are on track to receive an additional $2.2 million in non-dilutive funding from an STTR grant to support our single center retrospective outcome study at Northwestern in more than 8,000 patients, with plans to expand to a multi-center, multi-vendor outcome study with an additional 2,000 patients with our collaborators at the University of Colorado and Wisconsin. The NIH STTR grant also provides support for regulatory consultants who will help us get our first product, TCDFlow, through FDA clearance. We're using big data and AI to predict the future. Our vision is to combine aortic hemodynamics with anatomical information from CTAs and MRAs, as well as clinical information available in the electronic medical records, to create precision diagnostics and provide a comprehensive aorta risk score that can eventually be incorporated into clinical care guidelines. We anticipate FDA 510 clearance of our first product, TCDFlow, in 2026, followed by CBT code reimbursement and guideline inclusion. What we're developing is a platform technology. We're starting with thoracic aortic disease, as well as structural heart, including aortic valve disease. Future product development will focus on creating additional precision diagnostics for abdominal aortic aneurysms, as well as atrial fibrillation. Our competitive advantage is that we have exclusive access to the largest 40Flow database in the world. We have strong IP, and we have strong academic and clinical partnerships, which allows us to have an accelerated pathway for piloting and rolling out our technology. Most importantly, we have a strong team dedicated to achieving this vision. I received my medical training at Northwestern and my MBA from Kellogg. Gabby is a current MD student at Northwestern, and Dr. Markle and Dr. Allen are recognized globally as being pioneers in 40Flow MRI. Please join us in improving the standard of care for patients with aortic disease. This technology can save lives. Thank you. And now I will invite Dr. Michael Markle, one of our co-founders, on stage for the Q&A session as well. Great. Thank you. Turn it over to the judges. Excellent presentation. I'm an NU med grad as well. So my first comment is that wall stress has not been prospectively validated. You're using that as a biomarker, and your data is all retrospective. It's very nice to think that that initial case that you presented could have been saved had you had this information. But you're going to really need to validate this prospectively in order for us clinicians, cardiologists, and surgeons to put that in the guidelines. And hey, before you answer, I'm, as a judge, going to take prerogative here that we did not allow anyone else to have two people on stage. So I think the presenter should be the only one and answer the question. Sure. I can answer the question. So I think we completely agree that we're going to need prospective validation and studies after our multicenter retrospective study. And I think our timeline for getting the multicenter and multivendor retrospective outcome study completed is pretty quick because we can apply this tool retrospectively. And then I think, you know, if you look at the literature, there is good data between supporting the use of aortic hemodynamics to predict outcomes. However, the sample sizes are relatively small, like in the hundreds of patients. And so, well, we will have to do a multi, we will probably have to do a multicenter study. Could I go with the question? What's the predicate for your 510K? We, we have identified a few potential predicates and we are going to talk to our regulatory consultants to... So you're not at pre-subcube meetings or anything like that? Not yet. Okay. Got it. So with full disclosure, I'm actually involved with Clearly. And you're, you're going towards the path of what Clearly is trying to do with CT. With that, what kind of scans do you need? Do you need a high resolution scan? Or do you, any regular CT scan would do? Do you need certain, you know, heart rate control in order to get the best, you know, scan visibility? I'm just looking at the specifics, because, I mean, you haven't gone into the details yet, but there's a lot of requirements in order to get the optimal data to get, you know, the best results. Absolutely, I think right now we're looking, right now our data looks great for MR angiography, and we're expanding to CT angiography, ECG-gated, and I think from there we will see how well we can validate this for other CT modalities. Lastly, who else is doing this? So no one else is doing what we're doing right now. However, there are obviously other players that, you know, clearly in heart flow or in coronary artery disease space as a CT add-on. You have other companies that are developing platforms to identify, like intelligent care coordination platforms to sort of link patients who are identified as having thoracic aortic aneurysms to the correct, like, surveillance. But we think this would be beneficial to us as identifying more patients who need imaging surveillance. Thanks for presenting. How do you expect to monetize this? Is this an acquisition by an imaging company? Are you looking for add-on reimbursement, which is common with AI, and if you are, what do you think is gonna be the evidentiary standard? Will you need a randomized trial? I think, so the way we're thinking of monetizing this right now is an image add-on, so pay per image. However, we're also exploring different business models such as subscription-based models or a plug-in model to, like, one of these more ecosystem players. In terms of the randomized trial, I think this goes back to the question on the perspective study, and I think that's to be decided. Okay, I don't see any audience questions either here. Okay, thank you, I appreciate it, yeah. So let's, the next presenter, Proton Intelligence. Yeah. Can everyone hear me? Yep, okay, all right. Okay, all right, so I'm here today to talk about Proton Intelligence, and hopefully the takeaway that you get is not necessarily about my company, but it's about potassium. How does this click, oh, there we go. So, yeah, I'm reality TV show star, Steve Pham, from New York Med. I, since then, have gone really downhill and worked only on medical devices. At Berkeley, I teach about medical devices. At Roivant, I led on M&A of medical devices. At ZS, I consult with biopharma about medical devices. And at Echo, my pride and joy, this digital stethoscope, I've built some of the algorithms you have heard about already today on this panel. I'm here to turn you onto a problem that hopefully you're aware of, but if not, I really wanna make you aware because this is the journey I see my patients ending up at. So if you take your chronic disease patient, they collect cardio-renal, metabolic diseases over time, you know, comorbidities like hypertension, heart failure, kidney disease. Now, we're really good at not just maintaining them, but also using disease-modifying drugs at halting the progression and the slide down for them. So these are your common things, RASIs, MRAs, you know, soon-to-be aldosterone synthase inhibitors. But what's lurking behind all of them is potassium. And this causes an anxiety, hopefully, you know, not just for cardiologists, but also for me because they end up with arrhythmia in my department on crash dialysis, sometimes sudden cardiac death at my reality TV show. So what our solution is is the very first continuous potassium monitor. Sorry, the formatting's a little bit off. The principle is essentially we're using a CGM form factor to insert an ISF, interstitial fluid electrode, just like you would with a CGM. The only major difference is that the chemical substrate detects potassium. So we believe that potassium is the invisible barrier to cardio-renal care. Let me make that case to you. Potassium is everywhere and nowhere. We don't have good data on this. We think that there's about, if you do sort of napkin math, maybe 1.4 million annual severe hyperkalemic events, but that's what we know of. That's based on billing and coding. We know that there's a lot more hyperkalemia out there. One of the ways we know that is if you actually look at studies that do hourly checks on serum potassium, you get fluctuations as much as 0.5 on a daily basis. So that's pretty crazy, right? Now, if you're practicing like me, the first question that I had was, what would I do differently if I knew that the potassium on a patient I'm about to discharge with 5.2 could fluctuate all the way between 4.7 to 5.7? Would that change the way I practice? And I hope, you know, the scary thought to you is that you're flying blind, right? So what about heart failure? What about cardiac conditions? In this particular population, again, scarce data, but for new heart failure patients, on the bottom right, 0.9% incidence known for hyperkalemia. You wait two to three years past the new diagnosis of heart failure, it gets as high as 39%. That's an Australia and Europe-wide study, so it's a high-quality study. So this is really going to become a barrier to GDMT, and we know this because, actually, if you look at the guidelines, it clearly spells out in, you know, I put the red boxes, but everywhere it says potassium, potassium, potassium, electrolyte checks for three out of the four major drugs. And if you look at the barrier section to the latest guidelines, this is the only clinical barrier. Everything else is a social, you know, DEI, other kinds of payer issues, but this is the only clinical barrier. And amongst the audience here is people from Story Health. They did a great publication about two years ago, a narrative description of what happens to these patients when they, which handle outsourced GDMT model, try to manage their patients on GDMT. And so for a 12-patient narrative, they had 62 changes in medications that were relevant to potassium. And they highlight, in this section on barriers, specifically lab tests. That's the only clinical barrier. And that means about five lab checks per GDMT. In some protocols, that's as short as 60 days. Some of them go up to about six months. And you can imagine the stress that would cause a patient, and this is a quote from leadership at Story when I talked to them about it. We have some patients that live in remote rural areas that have to drive over an hour to get a lab. If you had to do GDMT virtually, and the promise was, we're gonna do everything virtual, you never have to do anything. Oh, but this one thing you have to do six times maybe, five to six times, that's kind of a pain. So hopefully I've convinced you that potassium is everywhere and nowhere, that it is an inconvenient truth that is causing suboptimal care, and that it's the primary barrier not only to heart failure GDMT, but also chronic kidney disease management, which also depends on RASIs. Resistant hypertension also depends on RASIs, and now we're seeing very interesting stuff with MRAs and then aldosterone synthase inhibitors, which are gonna be billion-dollar drugs tied to potassium. And then interdialysis sudden death, and I don't know if this, yeah, okay, in the lower corner here, near and dear to my heart, also acute dyskalemia, of which I see about 30% of my population getting some kind of acute dyskalemia. So that's why this is near and dear to my heart. So it's time now, okay? We have to do something about this. And so at Proton Intelligence, this is what we are working on. We are basically bonding a ion-selective electrode. So these are basically commoditized, mass-produced bacterial voltage-gated ion channels. And we are bonding it to this, okay? So this is the electrode, just like a CGM works, and the process is called dip coating, where you dip it into the chemical substrate, it bonds to it, and then you put it into a CGM form factor and implant it on your skin, just like this. So just like CGM. We're not doing anything different there. So how's our performance? So 80% of the work for CGM or any kind of ISF sensor is done preclinical. That's because if you get your manufacturing processes to snuff, you will be successful post into clinical trials. So we've spent, so far, three and a half years perfecting the chemistry. So we have, right now, greater than 95% correlation on a dialysis simulation with zero lag time. We also have specificity to potassium. So sodium, calcium, all your other electrolytes floating around that might trigger an ion channel, we've proven that they do not trigger a voltage change in our sensor. Okay, why now? What's the rationale? Why wasn't this done five years ago? Why hasn't Abbott and Medtronic done this? First thing is the economic case needed to be made. We are focusing on cardio-renal disease because not only are there drugs that can actually modify the disease, but there are now CMMI and payer pressure to do something about these patients through value-based care companies. They are aggregating them. You have heard about these companies, right? These are your Strives, Interwells, billion-dollar companies or startups, whatever, they're paper billionaires, but they're doing something about it, right? And they're putting pressure on their partners, nephrology practices, cardiology practices, primary care doctors, to deploy innovative new care pathways. We've talked to these companies. They can't wait to get our technology into their offices. We think that that will unlock the CPT code. This is a long journey, though, don't get me wrong. This will probably take about five years or more to get to some kind of conversation about reimbursement, but realistically, we believe we can get paid pilots in about two years. We will be on-market, FDA-approved, in about two years. De Novo, Class II De Novo. The other very interesting area of opportunity for us has been with pharma. I've mentioned aldosterone synthase inhibitors. There's also potassium binders. AstraZeneca has a billion-dollar one, CSL, VIFOR as well, and we're also working with phase one studies to start having conversations. What can we do to help biopharma monitor? So our first in-human study hopefully looks like this. This will launch in about a month. We want to have 10 healthy subjects, flatline, negative control. Potassium, sorry, CKD, late-stage, also flatline, negative control, but also at the edge of the normal, and then dialysis, where you should have a wild swing in dynamic range. Why us? So we are a team of passionate serial entrepreneurs. We are mission-oriented. We have significant experience with skin sensors, specifically with CGM, and I just want to highlight one person here on our board is the godfather, the very first inventor of CGM, Bill Van Antwerp at Medtronic. Also, recent pass away, Dr. George Bacris, sadly, but he really did help us form our thesis. We are IP-protected. Nobody is touching us. This is why. All the other competitors are doing something different, and they think they can do it, but they can't. Hemolysis occurs with finger sticks. AIEKG doesn't work. Optical sensors only work on dialysis fistulas. There's a lot of people who don't have dialysis fistulas. And then, I can talk about this later, but Medtronic, Abbott, Dexcom, those are CGMs. They work on enzymes. They don't know how to work with ion-selective electrodes. So we are raising five million. If you want to change care, if you want to create a new category, talk to me. We judge this. Steve. Oh, sorry. Yeah, David. Can you, so is it just a single, I guess, chemical reaction that you're checking, or can you add, perhaps, even sodium, or some of the other ones you talked about, calcium? Yeah, so I barely talked about this. I actually, I talked about it with you, but not, full disclosure, I talked about this with Dave. Sodium, actually, we've already stood up a proof of concept in less than 48 hours. Can we put sodium on potassium? There's no reason we don't think we can't. Sorry, let me rephrase. We think we can, but it may have to be multiple electrodes in one introducer. So it might be two, three leads in there. We're a platform play. We think we can insert a CGM with potassium, which would be very interesting, because there's interesting insulin dynamics between those, and we think we can complete the whole BMP. Yeah, just a comment. I totally agree with you in that we're trying to take heart failure care out of the clinic, or with a lot of involvement with physicians. Making changes in even oral meds, post-discharge or just out in the primary care setting, it is about fluids and also with changes in especially sodium and potassium. So this is starting to make sense, but the next question is how often do you need to sample in order to actually make the appropriate intervention that will have a clinical impact? There is no way and no reason why this should be done on a second-by-second basis. I'll be really clear about this. This should be done at most on an hourly basis, and that's in the acute care setting, which we are going to, by the way. We have an SBIR now to go into the emergency department. So we will be doing that on a Q15 to Q1 hour polling. However, in reality, this should be about Q daily, in my opinion, or maybe twice a day. I think there's circadian rhythm dynamics that probably require us to process on a daily basis, but not report on an hourly basis. I think what clinicians want to know is what's the mean, probably, and what's the range? And that's it. I don't think, we don't want data overload here. Hey, Steve, that was a really nice presentation. So you're closing on a seed now, presumably for some of those clinical study milestones. You're gonna be on a long haul, as you said, to get to a category-level CPT code. What do you view as kind of your fundraising arc to get there? I mean, how much are you wanna go it alone versus strategics? Are you gonna start building out a sales team kind of before that point, or is there an immediate adjacency to your point to kind of integrate into a diabetes form factor? Yeah, so a couple things. So we've already raised about 4.5 million to date on seed. This is our second seed. This will get us to A in about 18 months. At A, we're going to basically be ready for IDE, for pivotal study. We've already had numerous discussions, pre-subs with the FDA. We know this is not going to be a 510K, right? So you're right into, this is a long journey. So, however, strategic's important. We've had a lot of discussions, and in fact, Dexcom is actually, has already committed. So part of that is because of macroeconomics. I think five years ago, if you ask most med device companies, they weren't taking strategics, or they'd take it later, right? The reality of the funding environment right now, we have to consider strategics. That doesn't mean we will commit to Dexcom, but the fact that they've committed to us is exciting. Okay, I don't, I see an audience question, maybe. You can make it quick, please. So, Adrian Hernandez, a heart failure cardiologist at Duke. Can you help me understand how I'm supposed to use this? Because daily potassium levels wouldn't help, especially when drugs have longer half-lives, and we don't change drugs every day. Yeah, so I think V1.0 is a three-day wear, okay? And I think we want to be in parity with guidelines, which is 14-day check, right? Some people will say three-day. Up titration, post three-day, post 14-day. If we're in parity with that, then you put a three-day on immediately after, and then you put one on in about two weeks. We don't necessarily have to report daily for you. If you want just what the equivalent, what we think would be the equivalent of a mean that would report out just like a Quest Labs diagnostics, that's what we would do. And I think that there's immediate value for our patients, because that means they don't have to go to Quest Labs, right? I mean, how many of you have a question about Quest Labs, right? I mean, how many of our patients have been lost to Quest Labs, right? If you want to be in parity, then I think this is the more convenient way to be in parity. You just send a patch home to them, just like you would an iRhythm or something, a Zio. Okay, I think we're running out of time here. So just a few announcements. First of all, thank you for all the presenters. Great, great job, and thank you for sticking to time. So thank you. Before the audience leaves, I just want to request. So obviously, the judges have a tough job here in terms of going through and selecting a winner. So they're going to be selecting first place and second place, but the audience, you have a role as well. If you go to your app, there is something called People Choice Awards. So pick your favorite company and use your app to put in your vote. So the winner of the competition will be announced during the lunchtime after the judges had some time to deliberate over it. Again, thank you for all the presenters. Much appreciated. Enjoyed the presentations. Thank you.

pitch competition

medical technologies

iCardio

CareLog

HyloMorph

Vesta

CardioDiagnostics

AI algorithms

TCDFlow

Proton Intelligence

continuous potassium monitor