My name is Brandon Vasquez. I am one of William Blair's medical device analysts covering medical devices, but also some animal health themes as well. I am required to inform you that if you would like a complete list of research disclosures or potential conflicts of interest, please visit our website at williamblair.com. With that, I'm happy that we have Ceribell's CFO here with us, Scott Blumberg. He's going to run us through a corporate presentation. We will then go to breakout room Jenny B after. I'll remind you at the end, and we'll do a little bit of Q&A there. Thanks, Brandon. Again, I'm Scott Blumberg, CFO of Ceribell. What Ceribell has developed is a novel point-of-care electroencephalography or EEG platform that targets a variety of serious neurological conditions in the acute care setting, so the ICU and the ED. We've accomplished this by developing a new form factor to make the EEG setup very quick and easy to acquire and overlaying that with Clarity, which is our seizure detection algorithm, which allows for bedside diagnosis and triage of neurological conditions. Since we started commercializing our product in 2018, we've largely played within the U.S. market for seizure in adults in the acute care setting. That's roughly a $2 billion market. Over the course of the past year, we've expanded in additional indications, including pediatrics, neonates, and delirium, which has expanded our TAM, almost doubling to $3.5 billion where it stands today. In 2025, we delivered $89 million in revenue and are currently guiding to $112 million-$116 million in revenue as of our last quarter. In Q1, we delivered $26.5 million in revenue, which is a 29% year-over-year growth rate, and we're operating at very high gross margins, 87% in Q1. Just to set the stage for our discussion today, we view our growth in three horizons. We're initially focused on becoming the standard of care for seizure detection in the acute care setting. We've made inroads to our longer-term vision, which is to create a new vital sign around EEG to detect a wider variety of conditions in the acute care setting. Beyond that, we expect to grow beyond the acute care setting to develop new clinical settings as well as brain biomarkers for a variety of conditions. I'm only going to cover the first two horizons today, starting first with seizure. Typically, when investors hear seizure, they think about epilepsy, and that's not the market we serve generally. We serve seizures in the ICU and the ED. Epileptic seizures are visible. They're often characterized by convulsions. They typically last a few seconds to minutes, and although it's a very serious condition, a single seizure is rarely life-threatening. Seizures in the acute care setting are almost the polar opposite. They're generally non-convulsive, meaning they're hard to see. The patients don't typically convulse. They can last hours or even days if not treated, and because of the aggressive nature of these seizures, they can be life-threatening. There's a wide variety of conditions in the acute care setting that can lead to seizure. This includes things that you typically think about as neurological in nature, brain tumors, traumatic brain injury, stroke, but also conditions that aren't necessarily thought of as neurological conditions, things like cardiac arrest and sepsis. Essentially, when there is an acute assault on the brain, seizure is an autoimmune response. As I mentioned, these seizures, up to 92% of the time, are non-convulsive, and thus EEG is the only means to diagnose these seizures reliably. Physicians often say time is brain for stroke, and the same is true for seizure. That's true for two reasons. First is that if you let a patient linger in seizure for a prolonged amount of time, the outcomes can get really bad. As you see on the left chart there, patients lingering in seizure for hours can result in a very high mortality rate or secondary brain injury rate that can be establishing seizures for the rest of your life, permanent memory loss, disability. Beyond that, the frontline medication, which is a large dose of benzodiazepine, typically, if not administered promptly, becomes much less effective. You can see on the right chart that there's roughly 80% effectiveness of the frontline therapy if administered within half an hour, and that cuts in half if you wait just two hours. The societies have started to appreciate and adopt guidelines around the reality that seizures need to be diagnosed and treated promptly. The first society to adopt such guidelines was the Neurocritical Care Society in 2012, which suggested that if a physician suspects status epilepticus, which is terminology for a seizure lasting for five minutes or more, that EEG should be administered very promptly within 15 to 60 minutes. Over the past six years, some of the specialty societies, including the American Heart Association and American Stroke Association, have adopted similar guidelines appreciating the need for rapid EEG, and that's just simply not possible with the standard of care. This is a conventional EEG machine. This technology has been around in some form for literally 100 years. 2024 was the centennial of the first conventional EEG, and it was developed for the diagnosis of epilepsy in the outpatient clinic, and it's very good for that purpose. In the outpatient setting, speed is much less of an important factor. You can wait weeks for a diagnosis. It's very important to understand functionally where within the brain the issue is occurring for surgical preparation and other reasons. In the ICU, in the acute care setting, speed is the most important factor, and the precision matters less. This technology was applied to the acute care setting, and we think misapplied, and is really not set up to operate well within this setting. That's because of inherent limitations of conventional EEG. First, it requires a specialized EEG technician to set up a conventional EEG system. They have to manually part hair, clean the scalp, tape or glue on dozens of electrodes. It takes 30-60 minutes to set up. Beyond that, you need the tech to get there. You need them just to bring the machine in and get to the bedside. Techs are nationally in short supply. Most hospitals have tech staff 9:00 to 5:00, Monday to Friday, so it's hours of delay before you get a tech to the bedside. That's just for signal acquisition. Once you get the signal, you need to interpret it, and it takes a specially trained neurologist to interpret an EEG signal. They're not always available immediately, and when they do interpret, they don't continuously monitor. This results in something that is completely misaligned with what patients need, what the guidelines suggest. It results in hours or days of delay to get an EEG signal once a seizure is suspected. Once the signal is acquired, it's very rare that people are monitoring on an ongoing basis. That's where we're changing. We've developed a system that allows for rapid signal acquisition and continuous monitoring. Our system is very simple to set up, takes about five minutes to set up. It can be set up by a nurse and typically is set up by a nurse in the majority of our hospitals. Overlaying the system is AI, which allows for early detection of status epilepticus, and it continuously monitors the patient for the duration of the monitoring timeline. The system is shown on the screen, and the physical components of the system are composed of a reusable recorder, which is shown on the bottom left. Plugging into that is a disposable single-patient headband, which goes around the crown of the patient's head. It takes about five minutes to set up that headband, and there's an interface on the recorder that allows the nurse to have a feedback loop to ensure that the signal quality is good and that the electrodes are set up properly. Once the system is set up, it transmits the EEG data to the web through the cloud to an EEG portal, where the neurologist can promptly receive and review the EEG. Overlaying the full system is Clarity, the seizure detection algorithm, which allows for interpretation at the bedside as well as continuous monitoring. I'll highlight that here. The top left chart is raw EEG data. That's what neurologists interpret. What you see on the screen is 15 seconds of EEG data, and these tests are typically administered for hours at a time. In order for a neurologist to interpret raw EEG data, they need to flip through and review hundreds or up to thousands of pages of data. Clarity converts this very esoteric waveform into something that anybody can interpret. It does that two ways. First, on the bottom is the Clarity output for the entire recording of the patient's life cycle. What you see is seizure burden. It's the percentage of time that the patient was in seizure over a trailing five-minute window. I would infer that unless there's an epileptologist in the room, there's nobody in the room who can read the top waveform. I think we all can appreciate that when the bottom chart's high at 100%, the patient's been persistently in seizure, and when it's low and green, the patient's not in seizure. The other way the device translates data into actionable information is via alerts. On the top right, you see the recorder in red. When a persistent seizure lasting five minutes or more is detected, which is what physicians worry about most in the acute care setting, the device alarms and turns red. I'll bring it to life with a real patient study here. This was done a number of years ago at a small community hospital in the Bay Area. This is the Clarity output from about a five-hour recording. First thing I'll turn your attention to is the start time of 1:00 A.M. In the conventional EEG infrastructure, where most hospitals have tech staff 9:00 to 5:00, it would have been impossible to acquire an EEG waveform at 1:00 A.M. That already provides the value of having the nursing be able to set up at the bedside very simply. When the device was put on, the seizure burden curve immediately shot up to 100%, suggesting the patient was and had been for some indeterminate amount of time been in continuous seizure. The device alerted, and the staff was able to administer an anti-seizure medicine promptly. It didn't resolve the seizure. You can see a slight dip in the seizure burden, but remaining very high, and then going back to 100% seizure burden within 30 minutes or so. The care team was able to escalate treatment at that point and then resolve the seizure over the next hour or so. At about 4:00 A.M., seizure came back, change in medication, permanent resolution. This is a patient who spent some time in seizure, but without Ceribell, very likely they would have seized throughout the entire night. The EEG would arrive sometime during the course of the next business day, and the patient would have very likely had a very poor outcome. We're a very well-studied technology. We've got 150 peer-reviewed publications and abstracts. They demonstrate all aspects of the system, including the ability to change diagnostic care for the better, increase physician confidence, reduce length of stay, reduce transfer, things that matter both clinically and economically. From 2018 until just this year, we've focused only on the adult market. Our algorithm has only been cleared in patients 18 and up. Last year, we received clearance for the seizure detection algorithm on younger patients, both neonates and children, as well as a new form factor for infants. We just announced that we launched that technology in Q1 of this year. It expands our total addressable market by about $400 million, and we've gotten very positive reception on that technology. The reason is the clinical need. The core aspects of the clinical need are no different than adults. It's still seizure occurs. We don't have a means to diagnose. The implications of missing a seizure are very high. Because there's no baseline with neonates, you don't know what normal behavior is. It's even harder to see a seizure without EEG. The guidelines are actually much stronger in the neonate population. They say if the patient has an underlying condition that has a risk of seizure, put EEG on. It doesn't matter if there's clinical suspicion or not, you can't tell. You need EEG. The repercussions of somebody right at the beginning of life, when they've got years and years of life ahead, are profound. One hour in seizure alone can result in a 15% drop in cognitive and language ability permanently. The call to action here is significant. If you can promptly identify seizure in these young patients, you can change the entire trajectory of their life. I'm going to move now into the product and business model itself. Our business model is focused on two revenue streams. First is subscription. We charge a monthly subscription for access to our physical hardware, the recorder, as well as the software, both the EEG portal, where the information resides, as well as the AI overlaying both the portal and the recorder. That's roughly 25% of our revenue. Most of our revenue is a conventional razor/razor blade model, where the razor blade is the single-patient disposal headband, and that's roughly 75% of our business. Our business is really projectable. We've been commercial for about 32 quarters. We've had 32 quarters of sequential growth. You can see on the chart the projectable nature of our business, and that's inherent to the business model itself. The SaaS portion of the business model is quite literally a subscription, so it compounds as we add more and more accounts. The recorder, the headband portion of our business model, the disposable, is highly projectable. We've got a very high customer retention rate, and all this represents itself in high projectability of the business. We're in 680 hospitals now. We're in that sweet spot where we've got proof of the market need, but we're still only in 11% or so of the hospitals in the U.S. There's roughly 6,000 hospitals in the U.S. that have acute care services, either or both of an ED or ICU. Within our existing install base, our 680 customers, we're roughly 30% penetrated in terms of the patients who should be monitored for seizure. We have looked at that, both looking at patient populations using claims data, but also I think more compellingly, looking at the difference between our top customers and our average customers when controlling for opportunity, which bed size or patient inflows is a proxy for that. We see routinely across all hospital types, our top adopters do roughly 3x the volume of the average. Our strategy is two-dimensional in our core seizure market in alignment with the two dimensions of growth. The account acquisition side of our business really focuses on acquiring new customers. We added 33 new hospitals in Q2, 32 in Q1. We've got 5,400 left to go. We continue to expand within new hospitals as well as ancillary hospitals, including VA, where we got a FedRAMP High approval. We're the first medical technology, to our knowledge, to ever receive that level of cybersecurity clearance, and now access to children's hospitals as well by virtue of our clearance in both neonates and pediatrics. On the other side of the house, we drive growth within our install base, essentially three ways. One is by training new doctors, reinforcing habits, making sure that everybody within the ER and the ICU is aware of our technology and using it actively. Secondly is training, expanding to new departments within the hospital. We serve the ICU, the ED. Larger hospitals might have specialized ICUs, medical ICU, trauma ICU, neuro ICU, step-down units. Our strategy once we're into a hospital is land and expand and try to expand to other departments within the hospital. Finally is patient populations. We are often used for a subset of the patients who are at risk of seizure. Some of that has to do with awareness. There's varying levels of awareness of the many conditions that can lead to seizure, the conditions that were on that early slide with the iceberg on it. Our team is continually educating providers around various conditions that create risk for seizure. That's Horizon One, and Horizon Two will be much quicker. Horizon One is really taking what we've already developed. Rinsing and repeating, replicating our success and becoming the standard of care in the acute care setting. As we move forward, and we've made tangible steps towards this over the course of the past year, we intend to make EEG a new vital sign with the acute care setting. That has primarily to do with expanding into new indications that are impacting the same patients within the same call point. It can be leveraged very efficiently. It targets the same doctors, uses the same sales force. The first such indication we got was delirium. We received clearance of that in Q4 of last year. When a patient comes into the ICU or the ED, they often present with what's called altered mental status, which is simply they're acting in a way that's different than a normal baseline. That can express itself in confusion or not being responsive or slurring their words. A wide variety of conditions can lead to that, but the presentation is very similar across the conditions. The three things that physicians worry about most in the acute care setting are stroke, seizure, and delirium. We're expanding our platform on the same technology, deployed through the same sales force, to address these top three conditions. We're active in seizure. Delirium, we've got clearance and just announced about a month ago that we've initiated a pilot with intentions to launch that product formally in Q4 of this year and Q1. In stroke, we received breakthrough designation, which allows us to have an accelerated review process, as well as a path to favorable reimbursement. Delirium itself is about a $1 billion addition to our $2 billion U.S. TAM. Again, this is all domestic alone. We also announced on our last earnings call that we received a supportive proposal from CMS for favorable reimbursement for delirium through a New Technology Add-on Payment, which adds up to $2,100 per patient for use of our delirium technology. That is still preliminary. The final decision is expected in Q3 of this year, and if positive, will result in adoption of the new payment in October. I'm just going to drill into delirium quickly here. It's actually a much more prevalent condition than seizure is in the U.S. It impacts roughly 3 million patients in the U.S. About 30% of patients in the ICU suffer from delirium. Even more on the floor suffer from delirium as well. It really impacts older patients disproportionately. If you're mechanically ventilated, there's a very high chance that you have delirium, about 80%. Delirium's associated with significantly reduced outcomes. Every day that a patient lingers in delirium is associated with about a 10% increase in mortality. Even if you survive with delirium, it has long-lasting implications on the patient's quality of life. It can lead to dementia or dementia-like symptoms, things like PTSD as well. It's a very unpleasant experience that the patient carries with them long after their ICU stay. Currently, the standard of care for diagnosing delirium is a nursing protocol called CAM-ICU. It's basically a survey that the nurses conduct on the patient. Physicians have told us this is one of the only serious conditions that they deal with in the ICU where they just don't have a diagnostic and rely on a subjective measurement. When nurses conduct this survey, it's done typically once or maybe twice a day, and the output is binary. It's just yes or no. What we are developing, or have developed and are in the process of piloting and moving towards launch, is a delirium monitor that allows for continuous monitoring of delirium, allows for quantification of percentage of time that the patient had a positive delirium reading over a trailing window, and allows for better calibration of care by understanding how the patient's progressing as the doctor tries to rule and rule out the causes of delirium. It's essentially described as brain failure. It's organ failure of the brain. Much like organ failure of the kidneys or lungs or heart, you try to find the stressors that are causing the condition and then fix the environmental considerations in order to solve the condition. It's very synergistic with our current product. As I mentioned, it affects many of the same patient populations. The symptoms present very similarly. Roughly, not quite half of patients who have seizure in the ICU also experience delirium. Roughly 40% of patients who have delirium also have some sort of abnormal epileptiform activity, seizure, or seizure-like events. They're treated very differently. Anti-seizure medication can actually be a deliriogenic agent. If you're aware a patient has delirium, you might choose a different course of therapy. We think this is a condition where not just is there standalone value of the technology, but there's overlapping value such that we're able to add where the sum of the parts is greater than the individual parts themselves. As we move through 2026, our focus on delirium is to have a successful commercial pilot. We're live on that as of this quarter, intend to launch towards the end of 2026 or early 2027. I haven't mentioned stroke, upon securing the breakthrough of stroke, we need to conduct the remaining clinical and regulatory paths in order to seek approval for that technology. We're also working on our hardware. Our current hardware is perfectly capable of being deployed in these other indications, there's reason to believe that hardware modifications can accelerate growth as we make this a multifunctional brain monitor. Where are we going from here? We've taken this market from roughly $2 billion in 2025 in U.S. TAM. This year, with the launch of neonate, we are in roughly a $2.5 billion market. When we launch Delirium towards the end of 2026 or Q1 of 2027, it brings our TAM up to $3.5 billion. We've got opportunity to expand that even further with things like stroke, other new indications, or OUS expansion. I'll leave it there. We got five minutes remaining for Q&A, and we're happy to field any questions. We are the first and only seizure detection algorithm for neonates. We are the first and only for pediatrics, first and only delirium algorithm, the first to get breakthrough on stroke. I didn't mention this, but we've received FedRAMP High clearance from the government for use of our cloud in the VA system, which has a very high cybersecurity bar, which has been helpful both in expanding our TAM as well as being an indicator of the high standards of our cybersecurity program. With that, I'll turn it to Brandon if you have any questions or the audience.