Good day, everybody. I am Dr. Roy John, I work at the North Shore University Hospital in New York, and I'm proud to present the HRS Imagineer Technology Showcase. This is sponsored by the HRS Committee for Corporate Relations and Program Development, and I'm one of the members of the committee. And today we have Dr. Gus Bardi, who's the CEO of Bardi Diagnostics, who's going to present some data on the Carnation ambulatory monitor. Gus? Thank you, Dr. John, appreciate it. Let me begin by telling you that this is a company I started to honor my deceased wife, primarily because she died of cardiac arrhythmias, and Lorene died at the age of 52 from a multitude of rhythm disorders that were very, very difficult to decipher from a myriad number of external dermal monitors and implantable monitors. And so at that time, I was completely bewildered by the technology as to why I couldn't see my wife's P waves in order to diagnose various arrhythmias. So this is a company started in her honor. Just take you back 40 years, Paul Taboul out of Lyon first would teach his cardiology students the fundamental factor of electrocardiography in diagnosing arrhythmias, Recherche L'Empée. First thing you do is find the P wave. And that becomes a critical diagnostic tool. All of us know this who are electrophysiologists, this is stating a tautology, it's obvious. However, for at least 50 years, we've been looking at half a chest x-ray. And most people haven't been aware of this issue. And I certainly wasn't aware of this issue until it became a painful personal reality. But you wouldn't accept half a chest x-ray because it's the other half that carries home the point here. You can't diagnose myriad cardiac rhythm disorders from simply a ventricular signal. So why did I do this? Well, rhythm specific dermal monitoring simply has not existed until this development. It's essentially 50 years of inadequacy. And that inadequacy is founded both in technology and in woefully misunderstood developments on electrophysiology. In fact, probably wasn't until the mid 1990s, early 1990s, until we had a firm foundation on understanding rhythms. And now they continue to evolve. But the technology of dermal monitoring preceded that by 30 years and is grounded on a fundamental technological ignorance of the details of arrhythmias. So what are the learnings that I can share with you? First of all, the P wave. If you don't have a P wave, you really don't have anything with respect to rhythm monitoring. And that includes what everybody is currently obsessed upon, atrial fibrillation. There are a lot of pseudo-atrial fibrillations out there. The key here is low amplitude, low frequency content, and noise mitigation. And these are factors people don't really think about when they're developing these monitors. It's not filters and it's not frequency content. There are many things that go into being able to see the low amplitude and low frequency content. Second thing is EP level rhythm knowledge. That just simply has not been present in any company, and I'll emphasize any company other than here, which is pretty appalling given that we're monitoring rhythms, even things like ventricular tachycardia and pauses that are long, you would think are simple. They are not. No deep arrhythmia knowledge. There is no chance the report will be accurate. Third, recognizing that extended ECG recordings are like fingerprints, and this is something that's taken me years to recognize. You cannot subject these complex reports to some simplistic algorithmic approach. These are like fingerprints, truly. Everybody's rhythm is unique. Fourth, and this might shock anybody that's listening, the classical software used is nearly useless. It's useless primarily because it's essentially a 1960s thought process regarding the arrhythmias. Here's the fundamental problem. The people that read the rhythms that are raw ECGs that are pulled off of these monitors are at the mercy of the software. If you have an informed reader, they can't fight the software without an enormous amount of pain, and it's the software that needs to be dictated to by the reader, not the other way around. This is a fundamental flaw of monitoring technologies. And then fifth, as I call it, AAI versus AI, artificial, artificial intelligence versus artificial intelligence. To truly replicate the human brain from an electrophysiological perspective is not trivial. There's lots of garbage AI out there. There is no real AI, and this is critical, without tedious, detailed, high volume validation by highly paid experts, and that is painful to any company. The Carnation ambulatory monitor, I'll quickly zoom through because I realize we have time limits, but it's female-friendly, exercise-friendly, and shower-friendly, and I highlight that it's predicated upon 77 issued patents to validate the fact that this is new technology, and it detects arrhythmias precisely with clinically proven superior specificity. The CAMPATCH deliverables are as follows. The P-Wave, four levels of arrhythmia temporal context, first, 8-second standard strip with onsets and offsets, second, 56-second strip at 5 millimeter per second encompassing the arrhythmia onset and offset, third, 40 minutes of beat-to-beat RR plot. This is not a histogram. This is a beat-to-beat RR plot, 20 minutes before, 20 minutes after the strip. And then clinically valuable for drug mitigation and diagnostics is a 14-day continuous beat-by-beat RR plot. Specific rhythm diagnostics, in particular, I'd like to highlight, we do not use the amorphous term SVT. I have no idea what that means. We also have easy links between continuous RR plot symptoms and arrhythmias. At a recent point, where the RR plot and P-Wave integrity are manifest, at a quick glance, that 25 millimeter per second strip would look like sinus rhythm or sinus arrhythmia. The RR plot gives you a clue that there are two fundamentals going on, and if you look at the P-Waves, they're distinct. They are different because of the details provided in them. Another low-frequency, low-amplitude detail. Here's an example of a U-Wave being easily missed as a QT interval, if it's not brought out and highlighted as QT prolongation or QU prolongation. Just a quick example for those that know their rhythms. This is an obvious WPW block in the accessory pathway with ABRT kicking off during sinus rhythm. The CAM report structure is unique, it's uniform, and it highlights specific rhythm diagnostics. There's a uniform sequence on the right, summary on the left with some details about the patient. Then what that leads to is, if this is a blown up view of that, if a rhythm is present, the box is black. If it's not, it's not highlighted, but a physician can go quickly to look there to see whatever is founded. Case in point, here's an example of a patient that has three distinct arrhythmias. To my point, that everybody is much more complex than we think they are. The CAM visualization technology allows a continuous recording. Here's a patient with atrial fibrillation. You have a visual of that onset of atrial fibrillation in the evening, ending in the following evening. An example of the report where one can click, for example, if you blow up that one segment and you click on any of those flag buttons, strip 32, 34, 36, 38, the others are buried underneath. You can see where the atrial fibrillation stops and then restarts. This is not a sinus restart, it's a 5.6 second pause. Another example of the value of the continuous RR plot. Here's a patient with frequent atrial ectopy, sometimes block PACs, sometimes conducted PACs. If you look here, you see the RR plot telling you the time of day when the block PACs are occurring and when they're conducted, you see a different time of day. Now this could be medication related, it could be disease related. One doesn't know, but you have a visual of when these events are occurring. The data has been validated in clinical trials and publications, controlled studies. Here's an example of a Holter study published in the American Heart Journal in 2017, standard patch comparison. For those of you that want to dig into these things, that's published in also the American Heart Journal 2018, showing superior diagnostics, primarily the key finding here being different clinical actions. Case in point, a standard patch shows what looks like an SVT during atrial fibrillation or a fast afib rate, when in fact it's ventricular tachycardia. Here's an example where low frequency content doesn't just give you the P wave, doesn't just give you the T and U wave, it gives you QRS nuance. I will end with that because I think we are over time, but I'll leave this one up here as a concluding slide, and it's a strange concluding slide because it shows something that's quite unusual. I'll allow the team to stare at that for a long period of time to again, show the value of low frequency, low amplitude content in a patient who underwent heart transplantation. With that, I'm thankful for the opportunity to present this technology and look forward to questions. Thank you for that presentation, Dr. Bahrdi, that was excellent. I can understand your point about the accuracy of P wave measurement, that last EKG was fascinating, shows two P wave morphologies in somebody with a heart transplant. I think that's exactly what we're looking for when trying to diagnose difficult rhythm abnormalities. Now, one of the commonest problems that we encounter with these patches is the tolerability and acceptance by patients because a lot of these patients develop allergy to adhesives, find it difficult to keep them on, interferes with their daily activities, sweating, showers, all of those issues. Can you address some of that for us? Sure. The patch is designed with women in mind, as well as men. The sternal location is a wonderful location because it's less irritated, it's less movable. The sternum does move, but it doesn't move as much. There is perspiration issues there, especially on women with large cleavage, but it's not intolerable for the most part. That said, people that perspire a lot will not have the patch adhere as long, regardless of where it is on the chest. That's just a fact of life. Or the preparation isn't as robust, it will not adhere as long. For example, men with fuzzy chest, they need to be shaved cleanly, skin needs to be rubbed. People need to be tolerant of not acting vigorously and sweating in the first 24 hours to allow the adhesive to stick better, not showering right away. But these patches, they're all useful for showering and activity. And then ideally, many people can easily go 14 days, others cannot. Others get irritated. And that's just the fact of life and the diversity of skin. That said, it's certainly a lot easier to use than multi-lead systems. And certainly, probably the right step before an insertable loop recorder. At least some effort should be done. And then remember that even if it doesn't adhere for long periods of time, the diagnostic yield for serious arrhythmias progresses over time, plateaus roughly around 10 to 12 days for the vast majority of people. And yes, there are outliers, but they're the exceptions. Does that answer your question? Yes, it does. Yes, it does. So the other question was, you've now presumably done a number of patients. And have you encountered major issues? What is your tolerability for these things and skin reactions? Yeah, skin reactions aren't really that big of a deal. Because when people start getting irritated, and some do, especially people with fair skin, younger people that have sensitive skin, they'll simply peel it off. And they'll have some redness and irritation. But out of 160,000 uses, we've only had one patient who's had blistering, serious blistering. But that too went away. Most people will simply pull it off if it starts to irritate. That sounds good. What about the costs in relation to the available technology on the market? Is it all covered by insurance? Yeah, well, the costs are capped by insurance and Medicare. And so we all operate under that umbrella of financial restrictions. The ability to use these devices currently for extended monitoring is going to be revised. Medicare has revised some of those reimbursement rules. And those will be published, I believe, in July for beginning in 2021. We think they may be more reasonable. Holter monitoring, the one-day, two-day monitoring, has traditionally been severely under-reimbursed because of the reuse concept where you buy as capital equipment, the monitoring technology and the software. And then all you have to do is use the electrodes and put a series of wires up. That's no longer viable for most hospitals. That actually is much more expensive than paying for a patch or for allowing part of the billing procedure to be put upon the company, part put upon the hospital. That's called split billing, and that tends to work reasonably well for most institutions where the patch is not bought, but it's provided to the hospital, then the hospital will bill for application and professional interpretation, and then the company will bill for the technical interpretation of the recording. And finally, one of the issues we come across, especially with BioGel and some of the other competitors, is the false positive rate for atrial fibrillation. Have you studied that systematically to see if you have a false positive rate? Well, I think if you look at one of our publications, the second one, 2018 publication, it's not ... We helped sponsor it, but that was done by independent investigators. That one demonstrates just what you're touching upon, false positive diagnoses of atrial fibrillation when other arrhythmias are present. Most commonly, that is frequent atrial ectopy, frequent ventricular ectopy, confused by an R-wave detector as AF, atrial flutter with variable conduction, atrial tachycardia with variable conduction, even noise called atrial fibrillation. We have FDA-approved artificial intelligence on AF that, as I said, there's a distinction between artificial, artificial intelligence, as I like to call it, versus trying our best to take an electrophysiological mind and put it into the computer. And that requires, I mentioned this, and it's a serious component, high levels of knowledge doing boring work for long periods of time for many, many cases in order to truly get an EP-level AI on atrial fibrillation. And surprisingly, just imagine yourself in a room going over hundreds and thousands of strips denoting onset and offset of true atrial fibrillation, and then something else that could look like atrial fibrillation. As far as I know, we're the only ones that have done that. And it is painful and expensive, and you just can't wave a bunch of technicians in front of those strips to get that done. Or even cardiologists, you truly need electrophysiologists or highly trained electrophysiology nurses to do that kind of work. So the long answer to a very obvious question, there are a heck of a lot of false positives out there. And that has downstream consequences, poor downstream consequences for patient care. Great. Thank you so much. That was a great presentation, a lot of information, and looks like almost excellent technology. And I think we're all looking to see better delineation of everything else in these systems. So thank you very much. Let me take the opportunity to thank you from the HRS Corporate Relations Committee. It was wonderful talking to you. Thank you. Appreciate your time and HRS's time.

Carnation ambulatory monitor

cardiac arrhythmias

monitoring technology

P wave

dermal monitoring

ECG readings