Bill Plovanic, Senior Medical Device Analyst here with Canaccord. Welcome to our 45th Annual Growth Conference. With us next, we have the management of Ceribell. We have Jane Chao, Founder and CEO, and Scott Blumberg, CFO. With us, we're going to have a short presentation followed by a fireside chat. I'll hand it over to Jane.
Thank you, Bill. In five minutes, I'm going to give you an overview of Ceribell and the unmet needs we're trying to solve: our solution, financial model, and our pipeline. In a nutshell, we invented a platform for neuromonitoring that's EEG hardware, makes EEG acquisition very easy, and also the AI-powered algorithm. We call it Clarity. For now, we focus on seizure detection in acute care, that is ICU and ED, and that in the United States alone is more than a $2 billion TAM . Our future pipeline uses AI and EEG, expanding beyond seizure. We have breakthrough designations on delirium already. We're working on stroke and other indications. Our last quarter's revenue is $21.2 million. We are U.S. only and in 584 accounts. Year-to-year growth is 38%. Gross margin is 88%. The unmet need we're trying to solve is early detection and accurate detection of seizure.
Seizures are very common in the ICU. A third of the neurological patients have seizures, and the majority of them, more than 90% of these seizures, have no symptoms. You have to have an EEG to diagnose it. Very similar to stroke, often doctors say it's the second neuro emergency after stroke. Time to diagnosis is critical. If a patient seizes for a few hours, mortality and morbidity both go up significantly, as you see on the left chart. Treatment is readily available. Nurses can treat the patients if they know a patient has a seizure. Similar to stroke, if you treat the patient within the first, say, 30 minutes, the patient responds. If you delay that by a couple of hours, the patient stops responding to the treatment as well. Early detection, early EEG is the number one factor driving optimal outcome.
However, the conventional EEG, which was invented about 100 years ago, has been the gold standard. The problem with conventional EEG, without going into the details, is it takes the tech to set up. It takes very specialized neurologists to read. Overall, the hospitals usually have EEG Monday to Friday, 9:00 A.M.- 5:00 P.M. That's only 25% of the total hours. Getting an EEG always takes hours and very commonly takes days. Even when you have it, nobody reads it on a continuous monitor. Doctors are forced to make decisions: transfer, treat blindly, or just wait. That's the problem we solve. In a few minutes, our device would give the answer. Nurses can set up. Our AI algorithm would not get tired and constantly monitor the seizures. This is what the hardware looks like. The portal allows neurologists to read the EEG.
I want to walk you through quickly one patient's case so you can see how the device can help patients. First things first, at the bottom, you can see the device was set up at 1:00 A.M. If you go randomly to a big hospital in the neighborhood, usually you can't have an EEG at 1:00 A.M. As I mentioned earlier, even if you get an EEG, you don't have a neurologist read for you. In this case, within 10 minutes, the device said the patient had status. It sends an alert. If you see this little pink needle, that's the real bedside annotation. Nurses were able to treat the patient right away. In this case, the patient did not respond to the first-line medication. The seizure came back. They escalated the treatment. The patient stopped seizing. The seizure came back. They controlled the seizure again.
This patient would have seized probably the entire night. If you remember, a 20-hour seizure relates to more than a 30% mortality rate. It's not an outlier. It happens all the time. The clinical study we have published 35 peer reviews, close to 100% abstract now. Over and over again, we change physician decisions 40%- 50% of the time and significantly change and shorten ICU lengths of stay as well and reduce patient transfer. On the business front, you can see our revenue. In the company history, we don't have a single quarter that's lower than the previous quarter. Very strong and high growth. Our total TAM, as I mentioned earlier, is about $2 billion. That's 3 million patients and 600 hospitals in the U.S . Till today, we're only 3% penetrated. I want to spend one or two minutes on our future pipeline.
What we talked about is the current first horizon that has become standard of care for seizure management in the acute care setting. We are currently working on expanding to the pediatric and neonate population. The second horizon is to make EEG a new vital sign, still focusing on acute care. That is expanding seizure to stroke to delirium. Just like every patient who has chest pain has an EKG in the ICU, every patient who has altered mental status should have a brain monitor. The third horizon is going potentially beyond acute care to become a biomarker for other neurological or psychiatric disorder diagnoses or companion biomarkers for drug discovery. For this year, we are focusing on the first two horizons. The first horizon, we continue to drive account acquisition, as well as the same store growth as we've been reporting in the past quarter review.
We want to emphasize that we have developed two growth enablers already this year. The first one is we received a FedRAMP, enabling us to enter VA and potentially DOD as well. That expanded another 200 hospitals for us. The second one is we received the pediatric Clarity clearance in April. We are in the commercial pilot phase, and we expect to launch that product early next year. We're also working on expanding to neonate . Also, on the EEG front, we are planning to file for delirium 510(k) this year. On the product front, beyond all the software indications, we're also working on the hardware. We're potentially looking into adding camera, adding other features like quantitative EEG, as well as potentially alternative montage to the headset as well. With all this, we are very excited about what we have accomplished and more what's to come in the future.
All right. Thank you, Jane. That's a fantastic overview. Is that the fastest you've ever gone through the slide deck?
I've done it in two minutes, but not the deck. This is yes.
Let's start out with financials, and then we'll kind of walk through this. I think really surprised, you posted a really good quarter, and then you even raised guidance a little more than the Q2 beat. The stock traded off. I think a lot of us were the next day trying to figure out why. Any kind of general thoughts? I know it's an open-ended question, but it's the one I get more common than not for this stock.
It's both the one we get and the one we ask the most common as well because from our perspective, we hit our numbers, we raised guidance, and probably most importantly, we de-risked some of the tariff overhang related to China exposure. We were very happy with the quarter. Don't really have a clear-cut answer there. There hasn't been a specific narrative around it. We've got our head down, focused on execution, and we'll just keep delivering our numbers.
You mentioned the de-risking of the tariff with the Vietnam manufacturing. Just to clarify, you basically bought enough inventory from China to hold you through until the Vietnamese manufacturing gets up and running. Is that fair?
Yeah, accurate. When the China tariff pause happened, we started buying inventory. At the same time, we made the decision to go forward with a new manufacturing site in Vietnam. That'll be up and running by the end of this quarter, Q3. At that point, we'll have manufacturing lines in both China and Vietnam, which we like, the global diversity of supply chain. We'll likely run manufacturing through both sites. It also gives us the flexibility to shift part of or all of manufacturing from one country to the other to the extent one becomes much more attractive or untenable.
What gives you comfort that the new manufacturing facility will be able to get up and be as efficient as the one in China?
One thing that I don't think we talked about explicitly is that the manufacturer is actually one of our longstanding manufacturers with a second location. They're very familiar with the product. They're very familiar with us. They've got the training. They're sending staff over. We're sending staff over. We have a great deal of comfort with the quality. The plans that we established were not done haphazardly or quickly. We started thinking about supply chain diversity back in 2020 in the wake of some of the supply chain issues post-COVID. We were really waiting for the right time to execute. It was many months of planning, and we made a decision to move very quickly.
OK. I'm going to flip over to competition. I think that's probably the next biggest question I get. One of the kind of entrenched players in the market with the old technology is now coming out with something that's trying to compete with you. What have you seen in the marketplace? Any comments you can share with us? Thoughts around that in general.
Yeah. I mean, we created this category and have been the dominating player and leader here. Competition is also not new. It's important to emphasize that. Nihon Kohden is one of the conventional EEG manufacturers. They actually launched their point-of-care EEG device just a couple of years after ours, so that's about five, six years ago. We have seen quite a few attempts of competition, and we remain the dominant leader. The more recent ones, in many ways, are not a surprise because as we create the market, we see more coming. The reason we remain the dominant leader is multiple folds. One is our superior products, both from the hardware signal quality and ease of use, as well as Clarity, which is our AI algorithm.
We invested in the past eight years, but nobody was talking about AI back then, to build the infrastructure and in-house know-how, strong in-house team. Based on our knowledge, now our investors have in-house AI. Also, we received the breakthrough and therefore NTAP new technology add-on code also dedicated to Ceribell device. We really have the unparalleled clinical evidence, mostly published by Ceribell. We created this category that gives a lot of confidence and trust from the physicians and administrators.
As we think of the AI component, one of the questions we get, we follow a couple of other diagnostic-type companies, device diagnostic, that use a lot of AI. The example we get all the time is, that's great. Do they have data? Do they have annotated data? Where do you think you sit relative to anybody else trying to come in the market when it comes to that point?
Yeah. We believe we're significantly ahead. As we mentioned earlier, because we own the hardware, we actually accumulated more than 200,000 patients. That's millions of EEGs already. During the past eight years, we have invested heavily, have many epileptologists to label these EEGs as well. It's not just the size of the data. It's the quality of the label. The last one is in our user interface. We actually have feedback from the users. If users say, I disagree with the algorithm, they actually send us the feedback. We usually follow up. This takes us literally eight years to get where we are. Every six months, we look at potentially a new algorithm release. This was a dedicated in-house data science team. If competition doesn't have a dedicated in-house team, doesn't have another eight years, we believe that's very challenging to do.
I think one of the nuances for me in the presentation you gave today was just the topic of potentially including a camera in the device, which is, I think, something the competitor may be pushing on that you don't have today. How difficult, what does that require for you to include that camera into the device?
Yeah. Technically, it's very easy. The reason we have not included is we really want to optimize the balance of adding more workload to the nursing because it's the spirit of getting the device set up very easy. In most of the cases, you do not need a camera. The reason we're adding a camera is mainly because as we prepare for launching in the NICU for the neonate population, for that unique population, it's more critical to have a video to make sure the EEG reading is more accurate. As we launch that, we could selectively, when physicians want a camera, we have the optionality for the physicians. All the clinical data we have seen in the majority of the cases, you do not need a camera.
For us, lay finance people that pretend to be doctors, what's the advantage or the benefit of including the camera into the product?
For the neonate population, since I mentioned as an example, there are certain movements that can create a rhythmic artifact on EEG. For instance, if the baby is being patted, a rhythmic patting on the baby can look like a very rhythmic signal on the EEG. If you don't have the camera, sometimes you might read it as a false positive. That's one example to give you a sense of why the camera can be helpful. However, our algorithm is also getting very intelligent. For example, if there is a pump movement, the algorithm can detect artifacts more and more efficiently as well. It's both ways.
Perfect. All right. Let's discuss the sales force. I think one of the things that we talk about a lot and others talk about is just the fact that post the successful IPO almost a year ago, you were able to raise a little more than you expected at a little better price. With that, you took those proceeds and you started ramping up the sales force. I think it was like two, two and a half years faster than expected, right, pulling a lot of those hires in. The question I get usually is around productivity. As we think of new reps coming on and the ramp cycles and the productivity, is there any deviation you're seeing, either better or worse, from older classes of reps or prior classes? Is there anything different with this class versus others we should expect to see?
Short answer is not really. It typically takes about a year before a rep's activity funnels into the sales force. It's a few months to get trained, then the sales cycle, then the launch cycle. They're generally not visible to the outside looking in for about a year. Internally, of course, we see things much faster. When we say we expect an acceleration in account acquisition based on our investment next year, that's not based on hope or faith. That's based on the data that we see. We, of course, track progression of opportunities through a pipeline, track the number of champions, track passing through the purchasing committees. We also get purchase order data about three or four months before we launch. We have, with pretty high conviction, signals that the reps are progressing as planned.
Let me walk back on that for a second. When you report on the quarter, you had, I think it was 584 accounts, 585, 584. Are those accounts that signed six months ago, or are those the ones you just signed in those quarters and we haven't launched yet? How do we think about the timing for the next step on those? When do you actually book that? Because, like you said, if you could see into the future, are we even getting a glimpse of that yet?
The 584 is active accounts. Those are customers who have had Clarity turned on and are actively using the product. We get the purchase orders, as I mentioned, about three to four months before that. That is not reflected. The account base that we describe in our quarterly statements is in line with the RevRec. We have a pretty strong signal about what's going to launch well ahead of that.
Excellent. Let's talk about the commercial field force. You talked about the CAMs that you have. How should we think about that strategy and the ramp-up of that? What are their responsibilities?
The CAMs, Clinical Account Manager, can be thought of loosely as the acquisition arm. The Territory Manager is the acquisition arm. The Clinical Account Manager is the utilization arm. They come in right at about that launch time. There's a relationship handoff where the Clinical Account Managers start building relationships with the physicians and nurses. They jointly manage the launch with the Territory Managers. The Territory Manager steps away from the account, and the Clinical Account Manager focuses on driving usage. They really do that in three ways. One is through additional training. They train additional doctors, additional nurses, day shift, night shift. Second is expansions of sites of care. That would be other departments. Maybe we start in the ICU, and then we move to the ED or the step-down unit or specialized ICUs. Third is by educating on protocols.
The patients that our technology serves are quite diverse, and there is a diversity of understanding about the need amongst those patients. For example, the need for monitoring patients who have had prior seizures is crystal clear. Some other conditions, like sepsis, for example, might be lesser known. Our clinical team educates the hospital on the need, shares clinical literature and guidelines, and helps implement protocols. The hospital drives that, but they help educate to drive protocolization in order to drive usage to new patients.
I'm going to touch on the, you mentioned the account base. You kind of, it's a land and expand strategy with the ICU. You kind of go into the ED and maybe step-down ICU. Of those 584 accounts, just a rough estimate of how many of them are ICU only and how many have gone deeper?
Nearly all include the ICU. Many include some ED usage. Few are truly launched in standard of care in the ED. That is one of our larger opportunities, really doing what we did in the ICU in the ED.
When you go from that ICU to the ED, what's the magnifier of usage for an account? Is it like a 2x, 3x, 5x? How should we think about that? You're telling us that's the future and that's the opportunity.
Yeah. Maybe at the TAM level, very broad stroke is potentially half of the opportunity in ICU and half of the opportunity in the ED, just very high level, broad speaking. Of course, each hospital can be slightly different. Fully penetrated, they should be weighed equally.
OK. Let's switch over to the pediatric and neonate . You have the LMR in the pediatric. You have some early learnings from the neonate headband. Just discuss the opportunity in both the children's hospitals and the EDs for pediatric. How many hospitals are in today? Where's the opportunity? How should we think about this rolling out?
Yeah. To your point, the pediatric population, the biggest opportunity in our view is in the ED, less in the ICU. There are about 340, 115 million total ED visits in the U.S. Out of that, about 10%, 30 million, are pediatric ED visits. Seizure is the number one neurological abnormality that leads to the ED visits. About 20% of these pediatric ED visits happen in children's hospitals, which we are largely absent from because we don't have pediatric and neonate yet. 80% of these pediatric ED visits happen in the non-children's hospitals. Many of these accounts, of course, are our accounts. The limited market research or release we're working on is really trying to integrate this as part of our ED expansion as well.
To Scott's point, for the EDs we're already in, we can go talk with the physician, say, hey, how about your pediatric patient first-time seizure or patient altered have a risk of non-convulsive seizures? This is so new. For those of you who don't have the context, before Ceribell, the majority of the ED physicians probably have never ordered the EEG in their entire career. Now you are overlaid the pediatric, which is a more specialized population. We need to figure out the workflow, the protocol. However, the early learning is that there is an even higher emphasis on the neurological outcome of this patient population. There is plenty of research and studies showing that seizure burden leads to worse clinical outcomes for this patient population. We see a strong demand there. The neonate, there are 1,400 NICUs in the country. Many of them are in our existing accounts.
As we get the neonate Clarity cleared, we both can bond the neonate and pediatric deck together to target the 280 children's hospitals, which is not in our original TAM, and also expanding to NICUs as part of the departmental expansion as the same store grows that Scott mentioned earlier.
As we think of just your kind of growth into all these different areas, what's the biggest barrier to adoption? Is it just awareness?
I think it's a combination of awareness and just the time it takes going through different layers of the hospital approval process, especially in the account acquisition front. However, when we think about expanding to, say, different populations, that barrier in the existing accounts, we expect that barrier to be lower. It's much easier to go to the ED, say, expand to the pediatric population, or go to an existing hospital, say, expand to the NICU than acquire the entire new hospital or new hospital system.
It's a pretty audacious goal to make EEG another something we measure. What do you think we need to do to make that standard of care?
I think a big part of it is continuing doing what we have been doing with seizure and even more with delirium and stroke. What we did is not just providing a product. We developed a huge amount of evidence showing the clinical as well as the health economic outcome and working with the key opinion leaders. As we continue to do that in delirium and in stroke, we believe that it will change guidelines and continue showing the health economics. I think it's going to get to the pivot point that both in terms of number of accounts and using it, as well as doctors detecting disease state that they never expected through our device and the clinical guidelines.
Last time to finish up, you're sitting on a lot of cash. The stock is down. Just how should we think of use of that cash? Would there be something you could invest in to grow adoption? Buy back, I know usually you don't buy back stock a year after an IPO. It just seems like there's a dislocation between the opportunity and the stock price. Any thoughts? Has this been, you know?
Very much tied to the question you just asked. Number one is continue to grow our commercial infrastructure. We are only 3% in our seizure market. We need to continue to grow. Number two is we are investing significantly in our R&D and product to work on neonate , delirium, stroke, and other indications we have now disclosed. The third one is the clinical evidence. To make it a new vital sign, to make EEG a new vital sign to become standard of care, we need to partner and show the clinical evidence. All three fronts remain our investment focus.
All right. I think with that, we're out of time. Thank you.
Thank you.
Thank you.