Perimeter Medical Imaging AI, Inc. (PINK) Earnings Call Transcript & Summary

Good afternoon, everyone. Welcome to PropThink's Digital Conference. My name is Daniel Merrill, Editor of PropThink. I will be hosting today's event. PropThink Digital Conferences is a platform designed to give you access to subject matter experts, including prominent physicians, investors, executives and others with unique insights into the health care space. So before we jump into the discussion, just as a reminder for the audience that there will be forward-looking statements made, which are based on assumptions. We are not registered investment advisers, so nothing in this presentation should be taken as investment advice, always consult a professional before making any decisions. With that said, let's get into the agenda and speakers today. So we've invited 2 renowned experts in breast cancer and artificial intelligence to discuss the current landscape and ongoing trends in the space. The experts include Dr. Amelia Tower, a Board Certified Surgeon and fellow of the American Board of Osteopathic Surgeons and the American College of Surgeons. Dr. Tower serves as the Medical Director of the breast program at Texas Health Harris Methodist in Fort Worth as well as the Program Director of the General Surgery Residency. We're also joined by Adrian Mendes, the CEO of Perimeter Medical Imaging. He's a seasoned executive in the technology space, having built and scaled several companies. Most recently, Adrian was the Chief Operating Officer at Groq, an AI hardware company, where he was critical in scaling operations from a small team of under 10 employees to over 200 across the U.S. and Canada and also achieving a unicorn status with a valuation greater than $1 billion. So we will open the call with a roundtable discussion and close with Q&A. With that, let's get to it. Dr. Tower, Adrian, welcome. I've been looking forward to this discussion as it combines what I think are 2 very important and trendy topics. Adrian, let's start with you. So you obviously have a lot of commercial tech experience most recently at Groq, the AI hardware company that hit $1 billion valuation. Earlier this year, you joined Perimeter which is an early-stage commercial medical device company, quite different from your past experience. What about the technology, the data and the people drove you to make this move?

Yes. Thanks for having me, Daniel. Great question. So you're right, it is quite different to be in a medical devices company versus a computer and AI hardware company, but there's a lot of similarities. So when I first got introduced to Perimeter, I looked at it from the same way probably many of your audience look at it from an investor standpoint, right? So the first thing I looked at was, is there a big problem that this company is trying to solve? And the answer is of course, yes. A very, very big problem from a human level for obvious reasons. And then a very big economic problem as well. So there's a big problem here. Okay, great. Does the technology this company have -- does it address the problem? Do the customers like it? And as you'll hear from Dr. Tower today, the customers like it. The technology does address the problem, all right? So there's checkbox #2 checked. And then the next thing I looked at was the technology itself inside the company, do what they're doing? Or is it sort of a one-trick only or a little bit of a surface amount of technology? And the answer is no. There's actually a lot of technology within the company. The 2 major tracks are along the OCT track, which is the optical coherence tomography, which we'll talk a little bit about more. So an entire area of expertise there that the team has. And then also on the AI side, a very, very world-class team of AI. So it's got depth of technology that they're bringing together in a very interesting way. So I'm looking that from the outside. I then had the chance to meet the team. And what I found was a very committed team. The cofounders are still with the company since 2013. So 10 years still obviously very involved, full time employees here. And so we've got a deep team, good technology, big market, customers like the product. So for me, that checked all the boxes from a company standpoint. And then at a personal level, I -- as you mentioned, I came from a company called Groq, which is a hardware -- AI hardware company. I spent the last 7 years there, grew it from very, very small healthcare -- really from a very, very small company to the size that you described. And -- so that was wonderful. AI is something that's really -- I think, really going to change a lot of aspects of our lives. That was very deep in the technology stack, hardware all the way down to the semiconductor level. And this -- what this offers up is an opportunity for me to see firsthand, participate firsthand in AI, impacting everyday people lives in a very, very meaningful way. So when the opportunities presented to me, it was kind of a no-brainer by the time I did my homework. And since I've been here for the past 6 months, 5 months, it's been wonderful. All my sort of initial theories have been confirmed, and it's been very exciting.

That's great. That makes a lot of sense. So let's talk a little bit about the story itself. So study showed that about 23% of lumpectomies, patients must return to the operating room for a second surgery to get clean margins in breast cancer and breast ascites have called this the re-excision epidemic. And this question is for you, Dr. Tower as a Board Certified Surgeon with advanced training in breast vertical oncology, can you talk to us about how these re-excisions impact patients? And what strategies have surgeons such as yourself used to address the re-excisions?

Good afternoon. Thank you so much for this opportunity to speak on this amazing technology and discuss this further. So I think it's important right now to answer and understand exactly what a re-excision is. So a lumpectomy, which is also called a partial mastectomy is breast conservation therapy. That essentially means as surgeons, our goal is to remove the cancer with healthy tissue or what we call a rim of tissue around it on all sides. And whenever you think of this, think about a box. It's a 3D object, so there's six sides or what we call margins. What happens is that a pathologist will evaluate that tissue and then ensure that there is no further cancer left in that system we submit and remove from that patient. So that's signed in the postoperative setting after we've already completed the surgery, and pathology commonly can take up to 10 to 14 days to receive the final results. If there is cancer at the margin, then we have to talk to the patient and potentially -- and the standard of care is to go back to the OR to remove that healthy margins. So whenever we look at how a positive margin or re-excision impacts patients, essentially there's 4 ways that it impacts the patient. It's emotional trauma for the patient. It can be physical trauma for that patient. It can be financial trauma for the patient in the hospital that it can ultimately all of those things can also lead to delay in other adjuncts that are important for the overall prognosis and care of the patient. So I briefly want to go over those because it really is important to understand each element of this. It's all multifactorial, but it all comes together for best patient care and in oncology, it always matters. So whenever we look at a patient, the worst thing is to tell a patient, you have breast cancer. But it's equally awful to tell a patient, you have to go back to the OR for a second time. What that can do is result in emotional trauma for the patient. It can be undue stress. And up to 36% of those patients, actually, they decide not to go back to the OR, or I'm sorry, they go back to the OR, but they convert their operation, not from a re-excision, but to a mastectomy, which is complete removal of the breast because they're not confident that we can get the tissue we need with the re-excision. All of those things combined lead to emotional trauma that we want to avoid is surgeons. Whenever I say visible trauma, if we go oncologically at a patient to do breast conservation therapy, well, our primary goal is to remove the cancer, but we're also as surgeons trying to leave limited evidence that we were ever there and to preserve not only the functionality of the breast, but the cosmeses of the breast. So whenever we have to return to take more tissue, Oftentimes, there's postoperative inflammation that can also result in taking excessive tissue that otherwise we wouldn't necessarily have to take in the first operation had we known. That can result in disfiguring operations, and it can result in dissatisfaction of the cosmesis of the patient. And ultimately, that can also add to that emotional trauma. The financial trauma is significant. It cannot only affect the patient but the hospital. We know from our studies that every re-excision, it ultimately results in about $16,000 for costs. So that's an important consideration, that all these things, whenever we have a re-excision can also lead to delaying further adjuncts, primary cure for -- primary treatment for breast cancer is surgery that we work as a team to effectively reduce the risk of recurrence or it coming back and so we have to implement our adjuncts, which could include radiation. It could include chemotherapy, and it could include anti-estrogen or hormonal therapy, and these have to be disseminated on a certain time line. And so if we're delayed because we're allowing healing and second surgeries, up to 50% of those patients in the global setting, which is 90 days after the surgery, they can have further complications such as wound infections or dehiscence. This can all lead to delay of the adjuncts. All of that can effect prognosis. So while that is so important, we as surgeons have tried multiple ways, multiple strategies to eliminate all of that. we're trying to do our best for our patients. Some of that has been, well, a lot of surgeons routinely take what are called shave margins. That's where we routinely take additional edges or size that can result in cosmetic dissatisfaction and also removing healthy tissue that we otherwise would not need to. Some surgeons utilize other technologies such as ultrasound intraoperatively to get a better view. They also use additional technology that can use electromagnetic way to decide if there's potentially further cancer in the specimen or not. Those are not always reliable. They're not always effective. There's also ways that we can utilize our pathology colleagues. We can do what's called a frozen examination or touch prep. Unfortunately, that actually can compromise our specimen and what Perimeter OCT does is it doesn't compromise the specimen, it's nondestructive to anything that we removed. And we also sometimes utilize our radiology colleagues we do localization with wireless or buyer or needle localization, and that's effectively placing something preoperatively that requires another procedure that is invasive to help us find exactly where the tumor is and some of the technology can even give us information about the depth. Well, all of those things can result in essentially delay in OR. It can also result in additional trauma or emotional impacts to the patient. So whenever we look at all the strategies, unfortunately, we just haven't found one that works. But this using OCT intraoperatively, it gives you information and empowers the surgeon intraoperatively to make real-time decisions. It's nondestructive to the tissue and it ultimately can work and help reduce re-excisions which, if we can, in any way, reduce the fact that we have to tell a patient, you have to go back to the OR second time then it's an amazing technology that surgeons can potentially utilize and help with their outcomes and their best patient care.

So, Dr. Tower, you touched on some other imaging technologies that were currently being used. What are some of the gaps in these other imaging technologies that are used in the OR right now?

That's a great question. So there's a lot of new technology that's coming into the operative suite. As far as looking at the -- how reliable this technology is, how effective it is, how accurate it is and also how cost effective it is, all of that is still not established. So one of the things that also is a differing agent with this technology is, there's nothing comparable that provides the visionary structures of the tissue and the microstructures that this technology gives you. So the only thing that's comparable is essentially having a pathology colleague in the operating room with you. And so that's where this technology differs from other technology that's currently available and looking at essentially trying to figure out how we can reduce our positive margins.

Got it. Okay. So, Adrian, you mentioned while answering the prior question, you mentioned OCT technology. And that's been around for over 30 years and used in other spaces like ophthalmology, cardiology as well as other indications. Perimeter's technology is the first OCT imaging tool on the market for oncology surgeries. Can you talk about the differentiating factors of OCT, its benefits? And why it took this long to be introduced into the operating room?

Yes, absolutely. So as I mentioned earlier, there's 2 main technology tracks that our company is really investing heavily in, in terms of research and development. The first is OCT, the second is AI. So we'll get to AI later. But on the OCT side, so it's called optical coherence tomography. And so let me just give a little bit of a background on it. I'm an engineer, so I apologize if this gets too deep, but I'll try to keep it high level. It was developed in 1991 out of MIT. Dr. Fujimoto and Dr. Swanson were the inventors of it and sort of coincidentally, just last month, October, they both won an award. They were at the White House. They won the award for a National Award for Innovation and Technology for this technology. So it's -- and as Dr. Tower mentioned, it is used or perhaps you mentioned it is used in ophthalmology about 30 million times a year currently. The way to think about this is it's similar to ultrasound. So it's -- it emits -- no, ultrasound emits sound waves, emits waves, collects them after they bounce back against whatever is in the environment. And then from that, you construct a picture. Except what we use with OCT, what OCT uses is infrared light waves. What that gives is a couple of things. It gives a resolution that's much higher than ultrasound or x-ray which then allows someone observing the images to see at a much more detailed level what is happening inside of the tissue to Dr. Tower's point, it allows it's almost like you get that level of structure that you need when looking at an image. It penetrates 2 millimeters -- 2 to 3 millimeters into the tissue type. So versus an ultrasound which goes much, much deeper obviously. However, that's a perfect application for margin assessment, right? So you get the resolution you need and you get into that several millimeter range into the tissue. Okay. There's a science behind it, right? Then there's the engineering piece of it. So how do you take that technology and apply it into this space? So there were many engineering challenges. And the company was founded in 2013, so we're 10 years in right now. And many of the earliers were developed, not so much on the science, but more on the engineering piece of it of applying it to this application. One of the bigger -- there's many challenges. Some of them are easier to understand than others. One easy-ish sort of example is that OCT generally has a field of view of about 1 centimeter by 1 centimeter which is too small for this application. You need to be about 100 times bigger to have a product that's going to be useful here. So there was a significant amount of work figuring out how do you get the size of image that you need to be useful in this application. Another one, because the resolution is so high, having stability in the image, like you don't want a blurry picture, right? So you have to have a lot of stability. So a significant amount of engineering work closer to that. The flatness, you're looking at 2 millimeters in a very flat surface to take the picture on. So -- and those are just 3 examples of the engineer work. There's many, many other examples I could go on about. Okay. So a lot of work being developed has been done and continues to be done on the OCT technology from an engineering perspective. We got FDA clearance in 2021. So we've been in market since that point in time in a couple of years now. And you've heard from Dr. Tower, some of the feedback from those customers about the technology. And so we've been in operating room since then. I think that's the story.

It's -- that's good. I thought you were going to go into much more detail putting us to sleep on the engineering side of it at a high level.

[indiscernible] I don't have photons to look out.

Dr. Tower, you were a strong advocate being the first surgeon to use the S-Series OCT commercially in the U.S. what led you to your decision? And was there a so-called like aha moment?

There were several factors in me utilizing this technology as a fellowship-trained surgeon and also I'm a Program Director of a general surgery residency, so I'm helping to train the next generation of surgeon, I am a progressive surgeon. And I look for innovative surgical equipment and techniques to not only help and train new surgeons that are coming in to take care of future generations, but also to give my patients the best possible care I can. As breast surgeons were typically in the surgical arena, we are early adopters of a lot of technology that are advanced and help us to achieve our amazing logic outcomes that we do whenever we surgically manage our patients and provide intervention. So whenever I was given the opportunity to be educated on this, I was first enticed chiefly by the fact that it is something that I strongly feel and now is there's nothing else comparable to it currently, and it is meeting a significant deficiency that is in the OR whenever we undergo and we take on the challenge of breast conservation therapy for a patient, it is essentially providing us information that we otherwise have to wait for. There's nothing else that you can take a specimen and remove it from a patient and look at it in real time and see the micro-structures underneath. And by having that information that can either give you assurance or can give you invaluable results that can change or impact the patient perhaps having a second surgery to me, is priceless. And as a surgical oncologist is something that I'm going to utilize if I have the opportunity. But before I implemented it into my practice, I did have to have a lot of considerations because when taking any technology on board as a surgeon, you have to essentially weigh the risk and benefits. You have to essentially figure out how it's going to be integrated into your workflow of the OR, you have to figure out if it's going to be time-effective and if it's going to be accurate and work for you. And so by doing my own research, I essentially was looking at the studies and a case report of how OCT has been utilized in other surgical specialties. So that was not only reassuring to me that there were some benefits in the breast cancer arena, but also confirmed that I definitely wanted to participate and try this on my patients. And so whenever I did make the decision to move forward with this technology, the aha moment, there is all that it seeing is believing, really did happen and it essentially happened on the second case that I utilized this equipment, I was able to see something that was suspicious. I was able to intraoperatively make the decision to return back to the operating table why the machine is in the OR with me and within minutes, take additional tissue, rescan that issue, be confirmed that I was happy with the result, closed my patient and then the follow-up pathology correlated with had I not done that step, had I not utilize the technology, had I not turned around and taken an additional superior margin on that specific patient then I would have been having that conversation that I spoke of earlier. And so that moment to me, I guess, is my aha moment, but it really was thing is believing because as surgeons, we are some skeptical species sometimes. And so whenever we can see something that works, for us, with us and not against us, especially in the challenges of being in the operating arena and managing complex cancer patients then it's something that will definitely spend the time to utilize. So I guess that's what led me to it, and that's what made me a believer in that's what has sustained me in utilizing OCT for my patients.

That's great Dr. Tower. It really does sound like a breakthrough technology, especially for patients. I don't know if you've had any patient feedback because, I guess, during the surgery, they're not in there talking to you...

As surgeons, we don't like to talk to our patients or operate on. No, to briefly talk about that, anytime we're utilizing technology and processes that is included in our surgical preoperative discussion. So I did inform my patients, I was utilizing a new FDA-approved technology. I did not get the engineer version that Adrian gave, but I did give the surgical version of how this has been utilized for ophthalmology, for cardiology and hey, this is safe. One of the big things that's important with patients is letting them know this technology is not utilized inside their body. It's actually utilized on the specimen that is removed from the body and that's a big and a very important thing to disclose. And also that it's not destructive to the tissue. And so I informed the patient, the patients actually that utilized or I utilized it on, they have essentially that's included in their informed consent because I think that it's very important to be transient and let them know what we're utilizing in their operation because it's a collaborative effort to go to the OR. And so it was pretty -- it was very impactful. And having that conversation with that patient and showing the results of the pathology have not used this machine, had we not taken that extra tissue, what we would be talking about right now, unfortunately, you have to go back to the OR, not because I failed, not because we failed that operation, but because cancer never -- it never behaves by the rules. And cancer is microscopic in a lot of cases. And so this is something that there's no other technology that intraoperatively, you can utilize to see these ductile and microscopic tissue structures and that otherwise you would have to wait days that is essentially processed and placed on a slide for a pathologist to review.

Right. So, Adrian, we got the surgeon's point of view, we even have some feedback from patients. How are hospitals looking at this technology from an ROI perspective?

Yes, great question. So actually, the slide you put up is a wonderful one. So on the left side of the slide here is the current standard of care. And so what we're illustrating here is we just took lumpectomies as an example. And in the United States, there's about 23 -- sorry, 230,000 lumpectomies a year, with about 23% on average re-excision rates, that ends up being about 53 procedures that what I call are waste, right? Again, not because the surgeon did the wrong job because cancer is bad, it's evil. And so what that waste, when you dollarize it, times of $16,000 that Dr. Tower mentioned earlier, it's a lot of money. It's $850 million that we're wasting from a payer standpoint, right? Now imagine if you can bring that down into the single digits, right? So for illustrative purposes only, imagine you had it down to 3%, I incept, 23%. That's a lot of money that's potentially available to us. So taking my engineer hat on, put my business hat on, significant opportunity here. Even if we don't consider any of the traumatic the emotional and patient effects that Dr. Tower very eloquently talked about, there's significant savings to the health care system. Now we can provide these dollars are just for the U.S. only and also just for lumpectomy, is just breast cancer. So of course, there's great expansion beyond that, okay. When we look at that from the hospital side, so this is a payer. If I was a payer, this is what I'd be looking at. These are the savings I can have. There are some hospitals that pay. But as you know, in the U.S., usually the payer is separate from the hospital. So there's a couple of things the hospitals get out at this one is they get improved patient care, right? So as we start to move to a value-based care system, the ability to reduce re-excisions, improve outcomes, improve quality ratings, patient satisfaction scores. That's a very, very big deal to hospitals. So that's one element of it. The other element of it is sort of related, but it's the ability now that we're seeing 2021, right? So now that we've had this product in market for a couple of years, we're starting to get some good data from surgeons that are using it. And some of those hospitals starting to now think creatively about how do we market this, market our surgeons as being able to provide care to patients to potentially increase our top line revenues. So this is the view that hospitals are looking at this from an ROI standpoint. A lot about the R, the I side of the ROI, the most common business model that we have is where we place a hardware device at no cost into the hospital. And then we have a consumable element of the product use one per patient, one per procedure, and we charge for that. So it ends up being a relatively low barrier of entry from a hospital to get the technology in. And then as the surgeon uses the product, then they pay on a per use basis, which then allows the surgeon to get a feel for it to start to see some of the results. And it's not like they need to make a huge upfront financial commitment ahead of seeing some of those results.

Okay. Got it. That makes a lot of sense, especially on the number side, I think the way that you compare both the 23% re-excision rate and then 3%, there could be a lot of savings, not only for hospitals but just the health care system in general. So, Dr. Tower, this question is for you from these comments and testimonials, obviously, you and your colleagues have seen a clear benefit here. And can you talk about -- so when there's a new technology is introduced -- there is an investment of time to learn how to use the device and then also make the adjustments into your. Can you share your experience in implementing this technology into your operating room and workflow?

Yes. So anytime that you essentially take on new technology and your operative room, it's very important to have an idea of an effective way to implement it. But every surgeon has their own way to do everything. In preparation, there is somewhat of a learning curve with reading the OCT images. That being said, it is pretty quick. As breast surgeons, we do focus a lot of our surgical preparation and have been trained to read a lot of these images anyway. And so the learning curve is typically about after 5 or 10 cases that Perimeter staff comes in and helps read, helps work the machine. People feel comfortable doing it independently. Everybody's experience with that is different. That being said, there's also some learning -- some teaching module preparation beforehand and then just an atlas of images that you can review. So those are all advantages to kind of prepare where you can quickly review the images. To integrate this technology into the workflow in the OR was actually very easy. So it made sense to incorporate it. The typical process of a breast conservation therapy surgery is we essentially perform the partial mastectomy or the lumpectomy. So we remove that tissue. We perform what's fold and we orient it. So orientation is down with sutures. It's done with clip. It can be with radiographic clips as well. But in some ways, we denote the size, which would be the margins for the pathologist to review. And then we do an intraoperative specimen radiograph, which is essentially performed to document and for us to be assured that we removed the previous biopsy clip that is placed to diagnose the breast cancer and to also -- that we retrieved in the entirety any potential localizing apparatus if that was placed preoperatively. At that point, that's where it can be scanned. And again, it's nondestructive to the assessment. So you can scan 1 focused margin or you can scan all 6. I typically think if I'm going to use it. I'm going to use it. I use all 6 margins, why not? And each scan is about 1 to 2 minutes. So essentially, you're looking at 12-, 15-minute process. You can speed up by doing some other things intraoperatively as part of your procedure. You do not have to break but could see -- you can stay sterile or you don't have to break scrub. It is a process that while scanning, I prefer to do myself, you can also train the operative staff to do it, which has been opportunity to have other people assist you. So a surgeon can continue the operation that also in reading the images again, it becomes very quick and almost second nature. So the whole process is about 15, 20 minutes in its entirety, but it can be changed or modified to whatever surgeons workflow or patterns they have in the OR. For somebody like me, I utilize and trained medical students and residents so they can always participate in it. But it's an opportunity to integrate a technology that does not cause a lot of undue, I guess, training for staff. It doesn't destroy or struck any kind of the tissue that you remove or compromise anything. It doesn't add excessive time onto that operating clock that you have ticking. And it really just is a seamless process to consider and integrate, and it gives the surgeon opportunity to kind of modify it and individualize it into their workflow as they want to and then make adjustment.

Sounds quite easy, I guess, if -- so, Adrian, Perimeter is developing this next...

This is where you make a crack up to 30 years of medical treatment.

Yes, maybe we don't need to have read the intelligence changing.

Perfect segue into the robots taking over.

Yes, this is why we're moving into AI now. So the company, Adrian, is developing this next generation called the B series, OCT system, which uses AI. The -- there's an ongoing pivotal clinical trial right now, which is comparing the B Series against the current standard of care to assess the impact on these re-excision rates. Can you share some of the details of this trial? And what is the next milestone there, the readout?

Yes, absolutely. So let me back out them and just talk about the AI for a moment. So I talked about there being 2 major technology tracks within the company. OCT is one of then when we talked about that already. AI is the other one. We've got an AI team. They're world-class. We've got some world-class experts on our Board of Directors who are deep AI engineers and scientists and business people. So we've got a lot of depth of experience in our company. The nice thing about that is that a lot of people when they think about AI in medical imaging, they think about the image recognition piece of it. The piece of that for non-AI people that kind of draws the boxes around your friend's faces on Facebook and things like that which is kind of the base technology that is inside up B Series, which is going through the clinical trial right now, which I'll get to in a moment. But I just wanted to say a point first, which is that there is kind of approximately 4, maybe 5 other AI projects ongoing in the company that are outside of that image recognition piece that are deeper inside the machine that a health machine creates high-resolution images, helping the machine speed up the scanning, helping the machine take noise out of the images. So we've got sort of AI built helping our training sort of -- we talked a little bit about training. It's helping our training development. So we've got AI sort of like built throughout the company within our -- within everything that we deliver which is super important to what we're trying to do. Now to the B Series, what the B Series adds in on top of the S Series, which is the product that's in market right now, what the B Series has is an AI layer on top of it. And what this does is it will help the surgeon sort through all the images that are created, and it will highlight areas of interest. So it actually -- it doesn't make decisions. It's not a diagnostic tool. But what it does do is it helps the surgeon look at the areas that have the highest probability of being suspicious and helps them zoom-in quicker. There'll be 2 impacts of that. One is it should help just time in the operating room. And the second benefit of that is it does help expand what I'll say is expand our user base because it reduces the barrier -- the learning barrier to adopt the technology. So this will help from us from a sales standpoint, scale the technology even more and more surgeons and to get into more and more patients' hands. The trial is ongoing right now. It's -- at the current rate, we're projecting to finish enrolling patients in the trial towards the end of 2024. We have an opportunity to do what they call an interim analysis on the trial in the second quarter where we'll be able to look at the data and then make a decision whether we want to continue the trial to gather more data or if we want to stop the trial at that point and submit to FDA. So we'll have a better view of that in the second quarter. We've got 10 sites that are participating in the trial all across the U.S., the Mayo Clinic, MD Anderson, Baylor, so some very prestigious sites out there. And as we get more information from this, we'll continue to update before.

So the B Series is using AI predictive algorithms to assist surgeons in making real-time decisions. Where are you getting all this data from?

Yes. So with respect -- if we focus just on the B Series, or the image recognition portion, we've got a database of Dr. Tower referred to some of the atlas. We've got this AI atlas database, what we call it. We just got over 2 million images in it, which we used to train and to validate and verify the models. It's growing all the time. We've collected that from like actual samples, right? So we've got hundreds of samples from patients over time that we've collected. We've imaged them on the machine. So we've used those to create sort of a set of data. And then there are techniques you can use in AI to create even more data sets from that data augmentation methods. And out of that, we create many, many training set that's very, very deep. So we're at about 2 million right now. That continues to increase. We add to that all the time. Those are just on the breast side. And then we've got more in other places. So we collect this data in real time. We label it another area. We've got some AI activity going on is in developing -- in using AI to help the labeling, which is obviously a very labor-intensive exercise. So we've got techniques that are using AI to help with some of the pre-labeling. So this is -- I think this is actually one of the -- if I back out for a moment and just think about the company from a corporate valuation standpoint. When you think about AI is kind of a hot topic right now, maybe a little bit overblown in some manner, but there's fundamental value in it, right? So the market will figure out where that settles in over time, but you really do need it to move your business forward. There's 3 different areas that you sort of generate value with AI. One is the sophistication for algorithms, which we've got incredible algorithms. Two is your data sets. And this is a huge value for us. We're the only company that can create OCT images, which means we're the only company that has the ability to create this data set and has the ability to continue to grow the data sets. And we're the only company then, of course, that can label this data. So there's a huge amount of value in that, and there's a lot of different applications for that. Right now, we're talking about how we can use that data intraoperatively. And there's opportunities to move that also and to use that up at the biopsy level, possibly down at the pathology level, possibly. So a great expansion. Once you have the data, your mind starts to get very creative on how you can use this to help people do their jobs.

Right. And Dr. Tower, from your perspective, how would AI help you in your analysis and decision-making in the OR?

So in relation to this, I think the AI and utilizing all those specimens that Adrian just spoke of, these algorithms are essentially creating a big database of the common patterns and anomalies that you would see with breast cancer so that it can be easily identifiable with AI, which essentially, right now, I just described the workflow, you take the specimen, you scan the size and you're interpreting the images. So with the S series that's now in commercial use, the physician scrolls through the images. The B series will provide those suspicious images that the surgeon can readily look at and focus on. And not only am I interested to see and really just excited to learn more about the precision is going to add to the surgery, but I think that the speed that it assists with this already speedy technology. I think that that's going to really be exciting on our surgical forefront for breast cancer surgery, whenever we're doing breast conservation therapy because whenever we have those images right there for us to review very quickly and in such an effective and expeditious manner, I think that, that's going to help more people be amenable to adopt the technology and implement it and use. And again, this is all saving us from, it's not taking place of or eliminating the pathology. That's the gold standard of processing our specimen. So what it's doing is it's giving us a critical and invaluable information that we can grow and either reassure that we have a successful operation right then and there are that we may need to take additional tissue to ensure that we get a negative margin. And so that real-time information that AI, I think, obviously, will make it quicker is pretty exciting that we're just looking forward to the future.

It does sound like a very helpful add-on tool. I think a lot of people when they think about AI, they think it's going to replace surgeons or how...

Well, I hope not because I did. I need it done.

I think the patients think the same way as you do, Dr. Tower. I don't think anyone's comfortable having a doctor, sorry an AI running their surgery on them. So, Adrian, let's talk about the potential and the pathway for Perimeter's technology to be used in other specialty areas. We'll focus on cancer. So like we have lung, head and neck, prostate, colorectal and others. What is the total addressable market in these other indications? And what is needed to expand into them?

Yes. So if you think back to the previous slide that I had with the 2 pipe apps on it, right, there was 230,000 breast -- well, lumpectomies actually. If you look at this graph, it says 300. Well, there's 230 lumpectomies and the remainders are mastectomies. But that only represents -- you can eyeball this a little bit. About 15% of the total cancer procedures in the U.S. alone, right, which is almost 2 million. Diagnoses are happening every year in the U.S. alone. And the U.S. only represents about 10% of global cancer cases. So it's a massive market. And then the question is, from a business standpoint is how do you march through that market and bring this technology to those other places? So we know that margin analysis and getting the margins right isn't just applicable to breast cancer. It's really applicable to any type of cancer that you're going to exercise from the body. So we do have work, initial really early investigation work going on in multiple other areas. We're seeing some promising results. As we continue to develop those, we're starting -- like I mentioned before, I think we're starting to gather data images from these other tissue types. And I think one will end up guiding us as to where we expand next. So expanding into other tissue types is a definite. The question is which ones do we expand into. As you can see from this pie graph, there's 3 very big other areas, right, colorectal, lung, prostate, head and neck is very significant as well. And so what got that decision sort of where we see some of the easiest areas to identify where we've got an advocate surgeon in one of those specialties that is willing to work closely with us to help develop further in those areas. And then we'll go through. The good news is that our S series in market right now is a general tissue indication. So it can be used for any tissue type. So that's wonderful because we can be out in the market, starting to explore areas. In terms of the AI piece of it, which will eventually come, we will, like I mentioned, for breast, we've got 2 million-plus images on the breast tissue type. We'll have to create -- not create but grow our database in other tissue types to then be able to apply that to -- as we expand out these pie charts.

So clearly, there are many applications that OCT technology can go into and Perimeter has already exist in commercial efforts that also need to be funded. As of September of this year, the company had $18 million and you can be winning additional grants from this Biden Cancer Moonshot initiative that's on the slide here. Can you talk about how this initiative impacts Perimeter? And what will you be focusing on primarily within your budget?

Yes. It's a wonderful question. So when this initiative was announced in September, we were amazed. We're like, wow, this is a problem we've been working on for 10 years. We've seen the impact that this has. We have surgeons. We've seen patients. We all have probably have some in our lives that had to go back for a second surgery. So in one hand, it was like really nice to see the recognition that you get at the -- at this level, at the White House level that this is a problem that needs solving, number one. Number two, it's a problem that the government is willing to put money behind to help solve because there really aren't a lot of -- it's greenfield, right? We've heard that from Dr. Tower right now. There really aren't a lot of great solutions out there. And so from a government perspective, they're working really hard to try to help solve this problem. And that was a motivation behind funding ARPA-H and then funding the precision surgical intervention, PSI fund, specifically under ARPA-H. That funds got about $300 million assigned to it with the sole purpose of reducing second surgeries. So we submitted an abstract in September. That got approved. We were invited to submit a full proposal. We did that a few weeks ago. So now that's going through their selection process. I think there was 130-ish initial abstracts. That got filtered down to 20 full proposals that were submitted, so we were obviously one of the 20, and in Q1, I think February, some time, we'll get a decision. Okay. So our proposal is somewhere in the $15 million to $20 million range. That will be coming over 3 years or so, right? So that will help with the $18 million we have on the balance sheet right now. And the target for that money is really -- is to -- it's targeted -- it's nondilutive funding for one thing. So that's nice. And is focused on research and development. So this will really help, and we talked previously about area -- our 2 technology tracks. We've got lots of ideas on how to improve those technology tracks. The wonderful thing about this is it doesn't force us at all to move off of our strategic plans. The only thing it does is it allows us to accelerate those strategic plans by bringing more money in, allows us to staff up quicker, allows us to accelerate. So next tissue types, we'll be able to gather more samples quicker. We'll be able to develop the AI algorithms quicker. We'll be able to develop some more of the AI. I was talking about AI sort of throughout the machine to increase the resolution, to increase speed. A lot of those things we'll be able to bring to market much faster. There's some fundamental physics that we can work on in terms of the infrared sensors and light emitters that we will work on with the money as well. So it really helps in our R&D efforts. And so it allows us to focus assuming that we get this, we'll be able to speed up R&D efforts and allows us to free up some money that we would otherwise be spending in R&D and focusing more towards the commercialization side of the business.

Great. So, Dr. Tower coming back to you, it always just comes back to improving patient care, right? How do you -- just to tie everything together, how could this technology, especially with the AI add-on impact the standard of care in assessing margins during the surgery?

That's a great question. As a breast surgical oncologist, I can tell you that it's a privilege and a pleasure to be able to take a patient who has a life-threatening disease and to surgically eradicate it. And our goal is to always have the most successful outcome. And with technology like this, it not only enhances our precision with surgery that can reduce those 3 excisions and reduce all of those negative impacts that we talked about earlier, but overall, I think that it is going to -- it does help with the patient outcomes. And so with that incorporated and utilized as a standard of care and with the breast cancer arena, what that offers is a higher quality of care that surgeons can provide all patients. And what it also accentuates is value-based care with our ever-changing landscape of health care, we're definitely going toward. And so with the technology like Perimeter OCT Again, I think that it's one that there's nothing comparable to. I've seen the positive effect it has. It works. It can easily be utilized. And I think that we are going to have a lot more information in the future and with the AI component. I think that this is a way that is going to continue to reduce re-excision and continue to positively affect multiple facets of breast conservation therapy and the challenges that we've been trying for decades in the surgical world to eliminate. So I'm excited for the future.

Thanks for that, Dr. Tower. So we'll now start the Q&A portion of the event. You may submit a question using the text box at the bottom right corner of your screen. Let's just wait a few seconds. I think I already have a few that have been submitted. So the first question here is, I guess, is for Dr. Tower, how many surgeries have you done with OCT and an approximate how -- what has been the re-excision rate using OCT versus, I guess, standard of care?

So re-excision rates are different for every surgeon as far as how they're looked at and calculated in health care systems. I started adopting and utilizing the technology whenever I was coming to a new health care system, which was amazing advantage for me to be able to utilize it because otherwise, you're integrating your new surgical equipment and technology and essentially working with new pathologists and radiologists that was a really good opportunity to see how this could effectively go across the board. Whenever I started, I essentially was approaching around a 20% re-excision rate and with that, utilizing OCT, it immediately dropped to 0%, in fact, whenever I quit utilizing OCT routinely for all my surgeons just for a brief episode or time, I immediately had one re-excision. So it was confirming how valuable this was. So re-excisions are always different for every surgeon. Some surgeons won't admit though. But it is one of those things that also it depends on how cases are submitted because remember, I talked about earlier, whenever you talk to a patient and you counsel and you say, we got a positive margin. Obviously, we have to surgically manage this, some of these re-excisions essentially returned to the OR, not as a re-excision, which is a different CPT code, that essentially convert to that total mastectomy. So that's where sometimes, it depends on the health care system, how they actually essentially have the re-excision rate, but it definitely changed my re-excision rate. What is some of the other questions? Is this technology in Canada?

Yes, I can -- so yes, so is this technology in Canada. So the answer is no. We have FDA clearance. We're -- basically, we're focused on the U.S. market at this point. We've got FDA clearance. We haven't gone through the regulatory hurdles or whatever rings we have to jump through in Canada yet. It's on our list of things to do. Of course, our company was founded in Toronto. I was born and raised in Toronto. So I am Canadian. So I would love to bring this technology there for -- excellent. For all my friends and family that happen to be in Canada as well. And so we'll get there, we'll get there. We're just not there yet.

The next question, Adrian, is I think it's for the clinical trial that is currently being run. The question is, what is the current standard of care being used in the trial? And is the benchmark, the 23% rate that we referenced earlier.

Yes. So the current standard of care is -- well, I can probably ask Dr. Tower to describe current standard of care better than I would be able to from the outside. But basically, the way the trial works is the surgeon tells their standard of care, and then just before they're going to close, they open an envelope, and it literally is an envelope. And inside the envelope, it will say either device or control. And if it's a control arm, then the surgeon closes and that's the end of the procedure. And if it's a device arm, the surgeon takes the sample, the tissue and use of the OCT machine to scan it and then acts upon whatever they find. So that's ongoing right now. We don't have -- it's a blinded trial. Of course, we don't have data of how things are happening in the trial. The 23% is the national average based off of studies that have been done in the U.S. So we're expecting that, that's what we will see. But once we get the data from the trial later next year, we'll know.

So the next question, and this is for you, Dr. Tower, how are hospitals and physicians reimbursed for the use of the technology in the U.S.? Do the hospitals incur the cost of the re-excision? Or do they get additional payment for the second surgery by the insurance company? I guess how does coverage work?

So that's also can vary with systems and also with insurance. One of the things to know about this particular technology is whenever you're utilizing new surgical technology, CPT codes, which is what we use to bill our procedures, they're essentially called temporary codes. So oftentimes, they don't have any reimbursement that already I was able to get reimbursed for the hospital as well as with specific commercial for the surgical aspect of actually performing the -- utilizing the technology and doing the interpretation. So there are currently codes, and it's 0351T and 0352T and essentially adding the modifier 26, which shows that intraoperatively, I interpreted those images, and I acted with the surgical intervention based on that interpretation. So it is, I included an operative note and it's submitted and it has been reimbursed in cases. What was -- that's the question, right? Look, if there's another question now. When you say you used 6 scans, okay. Yes, that's a great question. So each -- every time you utilize it, there's one disposable that has used essentially $900. That's just that one disposable. So essentially, what happens is in the vacuum. And so you push a button and the vacuum sucks it down. So you can get a really good read and then release it. And then that the 6 scans is whenever you essentially are scanning each side, but you're utilizing the same disposable, which is really great because you're not having to use a bunch of equipment, and that doesn't increase the cost. So it's a -- one use of this disposable with the machine and not all surgeons and not all cases necessarily necessitates the 6 scans. Again, that 3 days specimen if you think about a box, there's top, bottom, front, back, side that's the lumpectomy. So again, I'm not really a betting person, so I have it. I'm going to use it. So I just do the 6 scans. And whenever I say scans, the scan take 1 to 2 minutes per side and then the images pop up, and you can actually be bringing or interpreting the images that say your superior margin, while your inferior margin, you release that vacuum, you placed it in, you hit the button, the vacuum hits again and bugs down and then it's scanning that inferior. So you can do a lot of it at the same time.

So the next question is also for you, Dr. Tower. What decision or decisions do you need to make? Or how do you think about or you need to do if you see an OCT image showing you a suspicious area?

That's a good question. So this is a biopsy proving cancer, right? And the purpose is, I'm utilizing the technology to intraoperatively get a real-time margin assessment. So if I see something that I deem suspicious, which I've done the training, I have to review the images in the atlas. I've done my cases. They got a number of that learning curve. If I see something say, for example, a dilated dug approaching the surface of the tissue, which is 2-millimeter slices, I see a dilated duct calcifications that again, I've practiced and read and feel comfortable calling then I will -- you can actually find exactly, very precisely where that is on the tissue and that's another advantage because the plate will project exactly where you find that area of suspicion, you can make a box on it. And right -- and in there, you can turn around and have a very precise re-excised area of tissue, and that can eliminate what we call a shave margin and a shave margin is essentially taking the entire aspect of that margin. And so it can lead to more focused and a more accurate area for revision. So that is another advantage that we haven't previously discussed about. So that's a great question. Use of that out for the area and double-check. That kind of is the double check. The double check would be, you can -- you have the option to utilize that vacuum again, and you can rescan if you resize tissue. And so I'm kind of one of those people that if I see something suspicious, on my specimen and I've re-excised more tissue. I'm going to stand that tissue re-excised until I'm confident that I got a negative margin and there's nothing else suspicious. And these structures are highly identifiable, these suspicious areas are very identifiable with training, obviously, from the otherwise normal or healthy tissue that you're scanning within this assessment.

So the -- I think the next question, Adrian, you might have answered already, which was the Biden PSI proposal was $15 million to $20 million, if I'm not mistaken, that you submitted?

Yes, that's correct. Yes. Yes, that's correct.

And then the question after that, what are some of the challenges that the S-series has faced in terms of scaling and capturing more of the market?

It isn't so much on the demand side, there's pull. There's customers that want this, obviously. We're -- in fact, as we continue to get more and more embedded in certain markets, especially Dallas where our U.S. headquarters is, we start to now get surgeons talking to their peers, and we're starting to get some great pull out of hospital groups and various surgeons, which is wonderful. So now it's more of an organizational, like how do we grow the organization in a way that we can support the demand. We've got a very deep pipeline in our sales pipeline, which is very encouraging. And so I think you're going to see over the next year and into the future, our sales really start to accelerate.

And then the next question, I think this is for you, Dr. Tower. The pathology is the gold standard, is there an ability to make pathology available in the OR? Or is that simply just not doable? I think it's the time maybe that limits it.

It's the resources and the time they don't necessarily -- there's other OR cases going on and a lot of health care systems don't have the resources or enough pathologists to commit to that, that another differentiating factor to consider is intraoperatively, they can do things like frozen section or touch preps that are quick about 7 minutes. What that can do is destroy some of the tissue, you're compromising some of the tissue because the gold standard pathology review, they have to do stains, let up process. So it still takes time and a certain process to do specimens, true specimen analysis. So with the Perimeter, it provides you that information, and it does strongly correlate with your final pathology. But again, you get it real time while you're in the OR.

The next question is -- so will the trial results, Adrian, for the B Series impact reimbursement coverage of the S series? Is there any read-through between the two?

No. So there are 2 separate products from the way the reimbursement schemes look at things from the FDA standpoint. It will help obviously, once we have S series coverage, it will help on the B series side as well. Because they are so closely related, so we'll be able to point to the results we're seeing in the S series to help the B series coverage. But they're treated as separate products from the FDA standpoint.

Got it. Okay. And I think this is the last question I see here interim analysis will occur in Q2. How confident are you in the study being positive? And if the study isn't stopped, when would final data be available?

Right. So we're hopeful, but until we actually see the data, it's going to be very hard to make a projection on whether it will be successful at that point or not. So, Dr. Tower said she's not a betting woman, I am a betting man. I'm hopeful. If we decide to continue past that interim analysis standpoint, it would be in the fourth quarter of next year that we enrolling all the patients, which would there's a few months of data gathering analysis needs to be done, which make the data available either towards the end of the year or early into 2025. So those are sort of the 2 milestones that we're running the company towards.

Got it. Perfect. Well, those are all the questions that I see in the queue right now. Adrian, would you like to add any final remarks?

I'm excited about what we're doing here. It's a very big market. We've got a technology that really seems to be making an impact. We've got customers that really like the technology. We're starting to get what I consider critical mass of our devices out there and information starting to flow back into us from use cases. I think that's exciting for the company. I think the fact that this is a promise being recognized at a national level is very exciting as well because that will give us some more win behind our sales. I'm hopeful that we will be successful in our grants, the ARPA-H grants. I'm hopeful that our estimated amount of money that we'll get from that, which is really just an estimate until we get the answer back will be close to what we are hoping for and that we're planning for. So we've got a number of very significant catalysts coming up here from a business standpoint over the next few months. So I'm excited. I know the product works, and I know it's having a great impact on our customers.

That's great. Thank you, Adrian. So, Dr. Tower, Adrian, thank you so much for your time. I think the event obviously proved to be very insightful. This concludes the presentation and the Q&A portion. The event was sponsored by Perimeter Medical. And if you're interested in getting in touch with the company, you may do so via the website, perimetermed.com. And if you're interested in learning more about PropThink, you can visit our website, propthink.com or follow us on Twitter @PropThinkers or handle. Thank you for attending, and I hope to see everyone again. Good afternoon.

Thanks, Daniel.

Thank you.