In modern medicine, a patient's outcome often depends less on the skill of their doctor and more on the speed and efficiency of the hospital system itself. When workflows break down, critical time is lost, and lives are put at risk. This episode provides a comprehensive analysis of the Viz.ai AI healthcare platform, a company at the forefront of solving this problem by using artificial intelligence to detect disease earlier and coordinate care faster. We are joined by Viz.ai's co-founder and CEO, Chris Mansi, a former neurosurgeon who saw firsthand how fragmented communication could lead to devastating consequences and set out to build a solution.
How can a simple software platform save lives in cases of stroke, pulmonary embolism, and cancer? In this deep dive, we explore the rise of Viz.ai, from its initial breakthrough in AI for stroke detection to its current status as a unicorn company installed in over 1,800 hospitals. We dissect how their technology optimizes AI in clinical workflows, not just by flagging critical findings on medical images, but by mobilizing entire care teams in real-time. The discussion covers the "time is brain" philosophy that drives the company’s mission and the powerful real-world evidence, backed by over 100 publications, demonstrating how this approach reduces treatment delays and improves patient outcomes. This has allowed Viz.ai to expand far beyond stroke to tackle aortic disease, hypertrophic cardiomyopathy, cancer, and more, proving the power of a unified platform.
We also analyze the company's evolving business strategy, particularly the crucial role of Viz.ai pharma partnerships. These collaborations with companies like BMS, Sanofi, and Medtronic are not just a revenue driver; they are essential for expanding the platform's reach into chronic diseases and ensuring that life-saving treatments get to the right patients. The conversation then shifts to the incredible Chris Mansi founder story, as the CEO himself recounts the patient story that inspired him to leave neurosurgery and become an entrepreneur. He shares invaluable insights from navigating the complex worlds of FDA clearance and CMS reimbursement, and what it takes to build a mission-driven company. Finally, we look to the future, discussing the biggest barriers to AI adoption in healthcare and Chris's vision for an "AI doctor assistant" that will augment physicians, reduce care variability, and democratize access to the best treatments available.
Introduction
Dr. Trevor Royce: Hey, Tim.
Dr. Tim Showalter: Hey, good morning.
Dr. Trevor Royce: Been a busy week in the Royce household. My son broke his collarbone, our first broken bone. We've got about 20 kid life years under our belts, and we finally had our first broken bone.
Dr. Tim Showalter: Well, statistically, it was bound to happen. How old is this one?
Dr. Trevor Royce: Six. Tree swing. Awkward fall, just a clean break.
Dr. Tim Showalter: That's tough. Been there before. That's hard. I knew something was up because you've got that 10 o'clock shadow going on.
Dr. Trevor Royce: I think we're up to 12 o'clock now.
Dr. Tim Showalter: Well, we don't have Paul, so it's just us. Should we get into it?
Dr. Trevor Royce: Let's do it.
Dr. Tim Showalter: Welcome to Health Tech Remedy, the show where three physician leaders in health technology tell the stories of new and established companies and interview leaders from the industry. I'm Tim Showalter, a radiation oncologist and prior medical device entrepreneur who is now focused on bringing AI advances to cancer patients.
Dr. Trevor Royce: And I'm Trevor Royce. I'm a radiation oncologist and researcher with experience in real-world evidence, informatics, and AI diagnostics. Missing today is Dr. Paul Gerrard, our fellow co-host.
Introducing Viz.ai: A Pioneer in AI-Powered Clinical Workflows
Dr. Tim Showalter: This week, we're talking about Viz.ai, one of the most prominent companies applying AI to clinical workflows. Their software is designed to detect critical conditions earlier and coordinate treatment faster, serving as an AI safety net across hospital systems. So Viz.ai was founded in 2016 at Stanford by neurosurgeon Chris Mansi, who actually did his training and initial practice in the UK, and he's a machine learning researcher as well.
Their first big breakthrough came in 2018 when they secured FDA de novo clearance for an AI stroke triage solution. So that's one of the first approvals of its kind and got a lot of notoriety based on that. From there, they've grown really quickly. They've expanded into cardiovascular disease by 2020. They've raised or reported nearly 300 million from top tier investors like Google Ventures and Kleiner Perkins and became a unicorn by 2022. Today, Viz is installed in more than 1,800 hospitals, including the majority of the largest integrated delivery networks. What stands out to you about Viz's early success and how they carved out such a strong position so quickly?
How AI Expedites Care for Stroke, PE, and Aortic Disease
Dr. Trevor Royce: It's hard to understate what they've been able to achieve on the product and regulatory side, the depth and breadth of their products. On the breadth side, you mentioned already their innovations for these AI products and FDA clearances, but I think they have over 50 FDA cleared algorithms at this point. And again, on the breadth side, that covers some of the biggest conditions that face acute care in the U.S. and globally in the world. Stroke, pulmonary embolism, aortic disease, intracranial hemorrhage, and I believe even cardiomyopathy through algorithmic analysis of things like EKGs.
Backing all that up, over 100 publications demonstrating the real-world impact, how these algorithms actually expedite care. So things like the rates of treatment for these diseases, shortening duration of hospital stays, and in some cases, advancing when these conditions get diagnosed by on a magnitude of years. So a lot of this, a common theme underlying all this is increased efficiency and care. And a lot of these conditions that we mentioned, there are very clear associations between timely receipt of care and outcomes. If you have a stroke, you want to get your therapy sooner, I forget there's some catchy ER phrase like time is brain or something.
Dr. Tim Showalter: That's it. Yeah.
Dr. Trevor Royce: But this is the exact setting in which more efficient expedited care through AI detection makes sense.
Dr. Tim Showalter: One thing I noticed on their website is that time is brain is literally one of their shared company values. So I think that really cuts at the root of what they're doing.
Beyond the Algorithm: Optimizing Care Coordination and Reducing Variability
Dr. Trevor Royce: Amazing. As a spouse to an ER physician, I've heard a lot about the speed at which it matters that you get these care. Tim, you've become quite familiar with this company because I think you've gotten to know their executive and leadership team over time in the health technology world, and you personally, living on this intersection of AI and clinical practice, how does this company stick out to you? What's important in terms of how they're publishing evidence, generating evidence, and ultimately convincing providers and hospitals and pharma to adopt what they've built?
Dr. Tim Showalter: That's a great question. And I think you know this as well from your work. I think ultimately clinical utility and patient impact is what really demonstrates value. And that's really across all stakeholders in the health system. So, of course, doctors want to take better care of patients. Patients want to have better outcomes. And for example, with the stroke algorithms, they want to be functional and their family members want them to be functional. Payers, of course, want to have value for the care that they're covering. And so really structuring their research to demonstrate that they're improving outcomes is what's so critical.
And so regardless of which customer you are in the healthcare chain, you want to know that these interventions are improving outcomes. And I think that's really the real insight and innovation that Viz has brought to the market is not just the algorithm. The algorithm is relatively trivial. It's about mobilizing care teams and coordinating care so that really good quality care can happen consistently and happen really quickly. And I think that's obviously key to pushing pharma to adopt and support Viz and how they're visualizing workflows as well and supporting the next phase of growth for Viz.
Dr. Trevor Royce: Is that how you see one of the major value propositions in their products is identifying areas in the healthcare delivery where time is of the essence. And by integrating these AI platforms, you can just move things quickly because you remove some of the subjective human thinking slowness of the system.
Dr. Tim Showalter: Yeah, I think time is of the essence is one of the main tenants. I do think there are other factors at play here. I think that consistency and so, getting, avoiding those major outlier terrible stories where it's great to have an average time or a median time of whatever the benchmark is, 15 minutes or something between the scan to the initial intervention for a stroke. But it's also really important to make sure that not only are you close to that target time, but that you avoid the one or two extreme outliers where you want to make sure that you're not having a handful of patients who are two hours until they're being evaluated.
And I think that's what the combination of an objective AI algorithm and thoughtful care coordination can do. We've both done a lot of health services research in the past, and I think that cuts at one of the tenets of reducing heterogeneity and variability and trying to drive better outcomes at the health system level.
From Acute Care to Chronic Disease: Viz.ai's Pharma Partnership Model
Dr Trevor Royce: Reflecting on the success that they've had and products to date and the health conditions, are there other disease areas that come to mind that maybe are ripe for this type of approach?
Dr. Tim Showalter: Well, it's an interesting concept. And if you follow their story, and we'll get a chance to hear more from Chris about what they're up to, but they've really got this application layer where they've got algorithmic support with care coordination. And what they've really pointed that towards in the past has been acute care scenarios where it's really critical to get things done quickly.
However, if you look more broadly within healthcare, there are also opportunities to provide insights at the speed of the individual clinical encounter, so to speak, and to really help doctors make consistently good recommendations for patients and make sure patients are getting the care they need in a timely fashion. So if you think about what they're doing in cancer and COPD based on the press releases, they're partnering with big pharma companies, BMS, Sanofi, Regeneron, Novartis, and they're teaming up with partners like Microsoft and Salesforce to give them the scale. And a lot of that is about embedding algorithms into enterprise-level clinical workflows, not just looking at radiology data, but also looking directly at EMR data.
And it's not the same value proposition that you might have for a stroke patient, for example, but understanding that the way the health system is set up where a lot of it is in that outpatient, a lot of the care is happening in that outpatient encounter and the shared decision-making, the documentation, the ordering is all happening then. Helping docs synthesize data quickly, be prompted or provided the best available information to make guideline concordant, best practice recommendations can move the needle forward. And my sense is that's where they're heading is to less urgently acute, but also still important and impactful applications.
Dr. Tim Showalter: So I guess to point it back to you, Trevor, I'm curious what your thoughts are about this pivot into the pharma partnerships. Obviously, there's a lot of challenges in the healthcare ecosystem for concepts like reimbursement and market access for some of these algorithms and other ways of driving revenue or essentials for revenue. What are your thoughts about this new business model for them?
Dr. Trevor Royce: Clearly, they have the machine in place to build these algorithms and develop the evidence. And they've applied that very effectively and demonstrated improved outcomes in the acute care setting. And as you mentioned, this idea of moving into other care settings, they have to think about how to take this beautiful machine that they've built, this algorithm development machine and find other areas and demonstrate the value. And the reality is, in these other areas, pharma has such an outsized pocketbook when we think about health technology industry and partners.
And we've seen this in so many different verticals of health technology. Obviously, with the real-world evidence and real-world data space, a lot of those products are also funded by pharma collaborations because they have the money and the hunger for these large data sets to do their development work and have these insights. And so it totally makes perfect sense that in fact, it would be reckless to not consider pharma again, because of just the large role that they play in development and these other care settings. If they can demonstrate a value proposition for pharma and what they're trying to achieve, I think it makes perfect business sense?
Dr. Tim Showalter: Yeah, I think as we move towards precision medicine and as things have gotten so complicated, you need to make sure that you have timely testing. That could be a certain PET scan or that could be molecular testing. And we all as oncologists or other clinicians, provide the right care most of the time. But there may be instances where there was one thing you needed to place that order or some barrier along the way. And I think this is where software like this can help make sure that the playing field is even and that everyone's getting the care they need.
Dr. Trevor Royce: Yeah, I think that's actually a really important concept to pull on a bit is this idea of personalized medicine or precision care really translates and encompasses across modern healthcare so broadly. And a lot of what we do in real-world data, real-world evidence, is driven by this idea of personalized or precision medicine, the right drug for the right patient at the right time. That's true also in the biomarker development world, where you want specific tests to demonstrate indications for specific patients. And then, of course, on the algorithm side, where you're trying to identify patients that need subservice before others or delay care for whatever reason, all that falls under this big umbrella of precision and personalized medicine.
Viz.ai's Strengths and Challenges in the Health Tech Market
Dr. Trevor Royce: So just based on our discussion, some very clear strengths that we've touched on are obviously their regulatory achievements and leadership. Clearly, they've built this machine in place for the algorithm development and regulatory approval, so over 50 FDA clearances, as we mentioned earlier, backed by strong evidence, over 100 studies. And then related to that is this hospital footprint that they've developed in the acute care setting for all these algorithm products. Now we're seeing these blue-chip partnerships that they're developing with pharmaceutical companies, medical device companies, and as you mentioned, Microsoft, Salesforce for scale and distribution. And then all of this has been supported by very successful fundraising. I think they've raised almost $300 million at this point. So they've got the bank account now to follow through on developing and expanding into these other areas.
Dr. Tim Showalter: It's interesting as they do have a very wide install base among health systems. And as they're expanding their services away from the acute care into oncology and in more medically treated conditions. They're going to increasingly bump up against some of the other players in that space who have also raised a fair amount of money. And as you point out, they've got a clear right to win based on many of those considerations. Generally speaking, for challenges with this, I think obviously hospitals and doctors are slow to adopt new technology, even when there's an improved outcome. So hopefully by embedding it carefully into the software system, they can get around that.
Obviously we discussed pharma as a revenue model it's essential to have them involved for these conditions they're generating the treatments obviously and those deals can be lucrative but they're very lumpy and that it's not an even pipeline of revenue often there's a lot of technical work involved and integrations with EMR and IT systems and there's some lift on both sides on the company and on the hospital side and they're well known within the field so obviously everyone's going to be watching them and so in an area where there's a lot of startups around they're clearly the incumbents now with a really broad footprint so people will be watching.
Dr. Trevor Royce: Prior success begets future expectations and so a lot of folks are eagerly anticipating how they continue to grow. So just to summarize, Viz.ai clearly evolved from a single-use stroke detection tool now into a broad AI-driven ecosystem and platform and development, just machine that touches now hospitals, life sciences and patients and everything in between. So it's going to be great to see how they can maintain their momentum, maintain their lead on competitors as they continue to scale and what's next for them. And fortunately, you should stick around for our next conversation because we've got Chris Mansi, who's the co-founder and CEO of Viz, and we're going to hear directly from him about how they intend to do that.
Dr. Tim Showalter: Yeah, it'd be great to have Paul with us for that conversation. So everyone, if you're a Paul fan, stick around. Chris, welcome to Health Tech Remedy. We're thrilled to have you with us.
Chris Mansi: Thanks for having me. It's great to be here.
The Viz.ai Origin Story: A Neurosurgeon's Mission to Fix Broken Workflows
Dr. Tim Showalter: Well, let's start with the origin story for Viz. So you founded Viz back in 2016, as I've read, while at Stanford. Can you take us back to the moment? What problem were you trying to solve? How did you land on the use of AI and this solution?
Chris Mansi: Sure. So I was a neurosurgeon in the UK and had come out to Stanford where deep learning was really taking off. Now you could tell the difference between a cat and a dog in an image. And it was becoming clear to me, playing with the technology, that you could also use it to read medical images. But for me as a neurosurgeon, that was less of interest. It was more of a how do you help patients?
And I realized from thinking through my neurosurgical practice that so much had to happen before a patient got to me, that no matter how good the surgery was, if the workflow prior to the patient getting to the OR wasn't optimized, the patient outcomes were much poorer than they needed to be. It was a real, real difference. If we knew about, say, a patient with a stroke or with a subdural hemorrhage or with a particular type of cancer in time and they got through the system quickly, outcomes would be much better versus if they didn't.
And the processes in healthcare, particularly in the hub and spoke model, are very burdensome. A patient will get a scan, and the scan will be read by a radiologist, the radiologist will try and get hold of an ED physician who will try and get hold of a neurologist, a neurosurgeon. We mapped it out and realized there were between 13 and 56 steps that needed to happen to make that happen. All of these steps can slow down, can break. The leading physician can get distracted by another case coming in. And that's why you saw such variation in care. So I realized that a consistent read of a medical image could trigger workflow. So it wasn't about making the radiologist more accurate or faster. It was about tying the workflow from when that patient first came in all the way through to eventual treatment across a variety of different diseases.
And we pitched this idea in a business school class to Eric Schmidt's Innovation Endeavors. He liked it. He was big on deep learning, obviously, and funded the company. And so my career went from neurosurgery to startup entrepreneur, which I think I probably didn't know at the time exactly what that meant. But I took the leap of faith and went for it. And we pushed forward. We focused on workload. We focused on the patient, took it through the FDA and then went from there.
Dr. Tim Showalter: That's remarkable. And I think just to comment on that, I think as clinicians in practice, so many of us just live with endless inefficiencies and we just assume that that's the reality that we have to live in and so it takes a visionary like you in 2016 to figure out there's probably a better way to do this. In the early days was there a particular story that you heard or experience that you'd had before that just really crystallized this need? What's the golden use case for this?
Chris Mansi: I remember a young lady who I operated on, she had an acute subdural hemorrhage. And the reason I remember her so well is the case went extremely well. And I was early in my career and the skin to skin time, I think was really good. So we were celebrating the operation. And 12 hours later, that young lady died. And you look back at the reason why. And it wasn't what we did in the OR. It was that took 30 minutes. It was the four to five hours it took to get her there.
And so it was a particular patient story that always stuck with me. I think we all have these cases where patients die and you're like, that shouldn't have happened, that stick with you. And from that, how could you have a system where people survive and thrive? And we ended up building a product to help us do that.
From Surgeon to CEO: Chris Mansi on the Leap into Entrepreneurship
Dr. Trevor Royce: I love the way that you describe that, how you had this moment where all of a sudden you were a startup entrepreneur and you hadn't really foreseen that. Was that existential at all for you as a physician and certainly as a neurosurgeon where it's famous for the rigor of training and the commitment that that specialty takes? Can you reflect on that?
Chris Mansi: Yeah. So during my neurosurgical residency, I had started a couple of education companies. The reason I went to Stanford was because I had a bug for entrepreneurship. I was planning not to leave neurosurgery, I was planning to take what I learned at Stanford back to the National Health Service of the UK where I worked. But I think just the combination of the technological inflection point of AI and the big problem that we saw. Actually, there's another part of the story, which is in stroke, which was our first use case, five randomized control trials in 2015 ended early because they were so clearly positive for a treatment called thrombectomy. And so you had this disease that is the number one cause of lifelong disability. You had a treatment for it. And I felt a huge gap that we could close.
And so going back to your question about neurosurgeon versus entrepreneur, I was on sabbatical from neurosurgery and in an environment where you're mingling with other wannabe entrepreneurs, with computer scientists, with physicians. I did this thing called the biodesign program, which really focuses you outside of your narrow focus as a physician, a surgeon, into thinking all of the different steps that need to happen to help this patient. And so I was seeing a different world, and it just became obvious this was something that I had to do. I still miss operating, and I love what my colleagues do, but I probably haven't had a moment in the past 10 years to look back and go, oh, is that the right decision? It's just been such a whirlwind since then.
A Company of Firsts: Navigating FDA Clearance and CMS Reimbursement
Dr. Paul Gerrard: I'd love to pivot and talk a little bit more about what's happened since you became an entrepreneur and about Viz. When I think about Viz, I think of it as a company for a number of firsts. Certainly one of the first, if not the first, to get a de novo for an AI device. One of the first, if not the first, to get a new technology add-on payment for AI and reimbursement for AI broadly. What have you learned from these things? What did you not know that it's probably good you didn't know? And do you think things have changed over the time since you had these initial accomplishments?
Chris Mansi: So, yes, it's true. We were the first to get FDA clearance for AI, first to get CMS reimbursement. And I think we have continued to be an innovator in this space. And the reason we've done that is our number one value is patients first, and we've always centered on the patient. So if the patient needed something and there wasn't a predicate device that would help that thing, that job to be done, then we would find a way through. And so we just took the right path versus the easy path. And I think that was the right thing to do.
What have we learned? I've learned that every three months, sometimes every week, you realize how naive you were the week before about how the FDA works or how reimbursement works or how, honestly, to hire people, manage people, all these different things. And so I've learned that the only thing that really matters is your rate of growth, both personal, but as a company, it doesn't matter where you are. And when we started the company, even though we were the first to get FDA approval, there were lots of other companies in the radiology AI space who had the technology. And so we might have been considered the underdog there but I think we felt that we actually had something that helped patients and something that also helped hospitals with their finances very significantly, so that dual value prop.
And so we're like okay cool, let's go at this. It doesn't matter that there's all this marketing out there about IBM Watson say for example. We feel we've got something that's really going to benefit people, let's go at it. And so we took the route, the DeNovo route. The received wisdom, for example, from bio design at Stanford was that if you can go 510k and get a predicate, you should do that. But we went to the FDA with that strategy and they were just like, no way. They didn't understand our device at first. And then within five minutes of meeting in person, they did, but they're like, you can't go down the predicate route. You have to do something new.
Heart sing moment, as a young CEO who's basically left his career in neurosurgery, you're like, oh, wow, that's going to take half a decade or something, and we don't have enough money to get there. But then you're like, okay, well, that's how do we work together? And actually, from that meeting to clearance was four months. Because we work with them, they work with us, we got the study done really quickly, and we got to submit it really quickly. And I think because they also wanted things that were good for American population, they understood that's what we were trying to do. There was just great collaboration. So I think that's probably one thing I've really learned. So growth mindset, but also people generally want the right thing, find a way to collaborate with them.
Dr. Paul Gerrard: That's great. I guess, do you have any tips on how early stage companies or early CEOs can really take steps that they can have that growth mindset in terms of both personal growth and company growth? We hear a lot of people talk about it, but connecting talk to action doesn't always happen?
Chris Mansi: I think, first of all, you have to believe to achieve something. And you can't be too clever and analytical about it. I actually remember some of my much more intelligent friends at Stanford, who maybe had backgrounds as management consultants, would analyze their potential business plan to the point where they'd never even tried. And the reality is so different. First of all, you can never really analyze all the things that are going to happen. The things that you can find out without going and actually doing it are limited. And actually the world changes as you go through it. So whatever your plan was a year ago, probably is no longer truly valid.
And so you have to believe it's the right thing to do and that you can do it and then go at it. And so, ironically, there's a little bit of naivety needed in entrepreneurship to say that this thing that companies that do the obvious thing never really win because the bigger companies will do it. It's the companies who find this product market fit that others didn't think was possible because maybe there's not an FDA pathway or people would have told us, oh, we need reimbursement and that's impossible for 10 years. Or people would say, well, hospitals don't have enough money to pay for these things. There could have been all those objections upfront or the technology is not good enough. The technology, of course, improves as you build. And so the world is changing. And so if you've got that belief and you drive forward, I think it's much easier to have that growth mindset.
How Viz.ai Chooses the Next Problem to Solve: A Mission-Driven Roadmap
Dr. Trevor Royce: Kind of reminds me of the timeless Marc Andreessen quote of strong convictions loosely held. It's like as an entrepreneur and a builder, you have this vision of your house, but you have to start somewhere. And your journey reflects that. I mean, you're certainly famous and made this impact with the first FDA approval and on the neurology space, but your portfolio of products and some of our listeners may not be aware of the extent of how much it's expanded. I think last I saw was over 50 FDA cleared algorithms, cardiovascular, pulmonary embolism, hypertrophic cardiomyopathy and beyond. So how do you guys think about that? How do you, in the big house of medicine, how do you know which problem to tackle next? Can you reflect a little bit on how you guys design your roadmap or what you focus on?
Chris Mansi: So our mission is to increase access to life-saving treatments. Okay. So let's unpack that a little bit. So first of all, there has to be a treatment that is going to help the patient. So if cardiomyopathy is a great example, we have an FDA cleared product for hypertrophic cardiomyopathy. Prior to MyoKardia and BMS having a treatment for HCM, it probably wasn't worth finding more of those patients and improving the workflow because they get there and their symptoms wouldn't improve. Once there's an effective treatment, all of a sudden that changes. Okay, so we look at areas where there's an effective treatment, but there's a big gap because if all patients are getting treated anyway, we're not going to have a delta on patients.
So we look there. Luckily, or I guess, unfortunately for the world, but also luckily for the opportunity, there's a lot of diseases where there's a big gap. Patients are still dying from cancers. Patients are still dying from strokes and getting really disabled from cardiac disease. There's hundreds of diseases where that's the case. And so while our first use case, I think it was very important that it was really time sensitive and had this really obvious value prop. As we've expanded, we've just seen more and more gaps that we can close.
Additionally, what we've done is we've expanded the input data that we use. Early on, because of the technology, deep learning was very good at imaging and less good at language. Pretty much every workflow, care pathways, every AI care pathway was triggered by an AI imaging read. Increasingly, like our work in prostate cancer or breast cancer, it's triggered by us actually just reading the EMR, summarizing it and pulling insights from there. And so now, actually, you can take much more multimodal complex data to trigger the identification of the gap. Maybe it's late-stage prostate cancer that really has it as a big issue. Maybe in lung cancer, if the patients aren't getting tested for EGFR mutations early enough, you start to get into more niche use cases where there's still this huge opportunity because there's the treatment at the end, but this care gap.
And so we look at those. Obviously, as a neurosurgeon, I have no rights to be an expert in lung cancer or prostate cancer. So what we do is we bring in experts across those diseases. And because we've become known for this workflow, AI workflow work, we have a lot of clinicians in our hospitals, in Mount Sinai, Cleveland Clinic, UCSD, who will come to us and say, oh, we see what the cardiologists are using or the neurologists are using. We'd love to use this for some use case I've never even thought about, sarcoidosis, let's say. And we go, oh, fantastic. And we look at whether we can actually both create a pathway around it and there's a willingness to adopt at scale beyond just that one user. And so, yeah, to answer your question, 50 FDA cleared pathways, but there's even more pathways that are EMR triggered pathways, which actually don't need an FDA clearance.
The Role of Clinical Validation: Proving AI's Impact with Real-World Evidence
Dr. Tim Showalter: That's great. And you can see the continued focus on clinical impact there and in terms of selecting which problems to go after. I think that's a really good point to maybe shift a little bit to how you and the team at Viz have focused on clinical validation. You've got more than 100 publications at this point and have really told a tight story about clinical validation and clinical utility and really thought with the patient and provider in mind. Can you share with us how you view evidence generation and what type of results you think are most compelling to your partners?
Chris Mansi: Yeah. So I think in clinical medicine, having published independent evidence is critical. That's what doctors basically look at to determine whether your software works or not, whether software device or drug. And so from the start, we've always worked with key opinion leader sites whose one motivation is treating these patients more effectively. Another motivation is publishing and presenting. And so there's hundreds of papers on Viz, and it depends on the use case. So always there's an ideally an outcome, like actual patient outcome based score. So in stroke, that's the modified rank and score of disability. But you're usually also looking at things like numbers of patient treated. Does that go up to fit our mission? Does the time to diagnosis go down.
We've seen in chronic disease, hypertrophic cardiomyopathy, we're taking diagnostic time down from two to five years to two to five weeks. So you're seeing a huge reduction in time to diagnosis and treatment. And then you're looking at other scoring mechanisms, like potentially time to get on treatment, time to surgery, particularly in stroke, that was a really big one. We call it your daughter needle time or daughter groin time. And so we really look at in the particular disease, we, the physicians look at in their particular disease, what determines good outcomes in their patients, and they measure that pre-vis, post-vis, and they publish on it.
Building an Ecosystem: The Strategy Behind Pharma and MedTech Partnerships
Dr. Paul Gerrard: You started to allude to this a few minutes ago when you were talking about Viz's mission on helping patients get access to treatments. But when I think of how Viz started, the company seemed to very much focus on partnering with the hospitals to ensure that patients got access to treatments at the time, at the point of care. But now Viz has started to work with some of the pharmaceutical companies in collaborations. How do you view these partnerships, and how does that fit into the long-term strategy for Viz?
Chris Mansi: So our mission, again, is to increase access to life-saving treatments. Who makes the treatments? The device companies and the pharma companies. So we've actually, very early on, we've worked with device companies. Our first use case in stroke, the treatment that really made the difference was a thrombectomy, so a surgical device that was made by Medtronic, by Stryker, by Penumbra. And so Medtronic was one of our first biggest partners and really helped push Viz throughout the whole, in nearly 2000 hospitals now. And so it's always been very natural for us to collaborate with the manufacturers of these treatments.
But I'd go even further and say that any healthcare company, particularly one like ours, that sits at such a key point in the workflow should work with, should focus on the patient, but should work with providers, should work with pharma and device, and also should work with payers. And we provide a lot of value for payers and that's a part of our business that's only just starting. But we have a lot of MA plans asking us, hey, it's clear that you improve patient outcomes. That's really good for us. How do we help incentivize hospitals to use this more? And so I think in terms of business strategy, it's about building that ecosystem of partners, all who want to do the best thing for the patients, and all who create and extract value from it. And depending on what value they get, we would ask them to be a partner in paying for the technology.
Dr. Trevor Royce: So is it fair, I guess, to say it kind of goes both ways where these relationships with the manufacturers, the pharma or the medical device, has accelerated the adoption on the side of care, the hospital side?
Chris Mansi: Hugely so, because I think hospital finances in the U.S. is interesting. They make a lot of money from procedural care. And so we found very high willingness to adopt anything that appropriately guideline-directed results in a surgical procedure, thrombectomy, a neurosurgical procedure, a cardiovascular procedure, because it helps the patient and makes them money. If there's no procedure at the end, there was actually a lower willingness to adopt. So think about hypertrophic cardiomyopathy. The treatment, actually, there is a surgical treatment, septal ablation or myectomy, but that's less commonly done. And for the drug, obviously the hospitals don't make any money. And so there, in order to get that information, program out and benefiting patients, we had to have a pharma partner to help us cover some of the development costs. So BMS came in and did that. And so I think you look at it use case by use case, who is creating and extracting value and you work with them dependent on that equation.
The Future of AI in Healthcare and its Biggest Adoption Barriers
Dr. Tim Showalter: It's clear that you're right at the middle of this dynamic time for AI in healthcare broadly. I'd love to get your thoughts on the broader picture of healthcare. We're obviously in a time of all these advances and it's really impacting patient care directly. What do you see as the biggest bottlenecks or barriers to AI adoption and moving forward to realize the vision of healthcare that we could attain?
Chris Mansi: I'll answer in two parts, because first of all, I agree, it's a really exciting time. I think 2016, 2017, it was really exciting because of deep learning, we could do new things. Now with generative AI, it's just exponentially increased what we could do with the technology. And partly through our work and others' work, AI is now just a part of medicine and hospitals adopted. So it's not like this crazy innovation that you have to make a leap for. It's something that you have to have. I would say most hospitals in the United States have an AI program. It's not yet fully adopted across all departments and all physicians. That's kind of the next step.
But I think there's this big opportunity now to essentially create the AI doctor assistant who, I think a bunch of you are physicians. What do we do in medicine? We look at tests, blood tests or images. We read clinical context from the medical record and we talk to the patient and get what their current problem is. And in our heads, we put that together into a summary of the patient and differential diagnosis, we make a decision. And when we have your talented senior residents or fellows, they do all of that for us. It is wonderful. And you come in and you make the decision.
AI can now do all of that. It can read the EMR, it can read the images, and we can listen to the doctor-patient consult and put that all together. And so I think medicine has got this huge opportunity to up level to being the people who really get to what the patient should be treated with, affecting the treatments quickly and democratically. So everyone benefits, not just the patient in front of you. So I think doctors are going to have the opportunity to massively expand the scope of who they treat and care for in their subspecialty.
But yes, to answer your question on challenges, The biggest challenge is right now there's tight finances in the hospital. And reimbursement, while there's been niche reimbursements, like our reimbursement for stroke, it's not broad. We need something like the HITE Act that drove adoption of the electronic health record for AI, because it's so clearly something that improves patient outcomes and actually reduces burden on physician by automated documentation, reduces delay in treatment, so it reduces cost, that actually there's an equation there where it just makes sense to double down on funding this technology so the hospitals can afford it. And so I think that's the biggest next barrier that we need to overcome. We actually just testified in Congress last week on this, and there's definitely willingness to improve reimbursement for AI. It's just there's an understanding gap of what it means and how to do it. And so we need to help government get through that.
Viz.ai's 5-10 Year Vision: The AI-Powered Doctor's Assistant
Dr. Paul Gerrard: Looking ahead five, 10 years from now, assuming that the infrastructure and maybe the funding to build the infrastructure is in place, how do you see AI transforming healthcare delivery and where do you want Viz to fit into that?
Chris Mansi: So, I think to continue what I was saying with AI doctor assistant, I think we expand what we're doing, the depth of what we're doing. So we've been historically known for an AI safety net and care pathways. Now, we're summarizing the EMR, we're incorporating ambient listening, we're doing a lot of that AI doctor assistant. I think in five to 10 years, you'll find that that's available for every single disease in the hospital. And it's always been my dream from the start.
Every single patient and every single test and every time you see the hospital, it's enhanced by AI. So you reduce the variability in care and get everybody on guideline-directed pathways. And I think that's easily within our reach in the next five, 10 years. On top of that, we do a lot of work with pharma, but are starting to do some work with biotech as well. I think speeding up how quickly we bring new treatments to market, so that for things like Alzheimer's and Parkinson's disease and different types of cancers, we can dramatically increase the effectiveness of our treatments. I think you'll start to see that. So there'll be more treatments, maybe there'll be more precise and targeted, so you'll need better precision medicine. With AI, that's reasonably easy to achieve. So I think you'll have a system where a patient comes in, they very quickly get diagnosed right into the right specialist and get on the right treatment much faster than ever before.
Leadership Lessons from a Unicorn CEO: Growth, Team, and Focus
Dr. Trevor Royce: I'll chime in and pivot a little bit as we approach the close-up here, just to kind of reflect on your own personal journey. You've gone from a neurosurgeon now to founder and CEO of an incredibly impressive health tech company, a unicorn. We'd love to hear a story of your reflections on your own leadership journey during that and how your view of that has changed?
Chris Mansi: Yeah. I think luckily very early on, I realized that as a physician, you didn't know everything and you actually knew very little about how to get companies started and all the different things you need to do. And I developed very quickly a respect and appreciation for all the other things, skill sets, whether it's sales or marketing or product management. All of these are disciplines just like neurosurgery or oncology that it takes years to master. And so what I've learned on top of what I said about the importance of growth and always having a growth mindset with you and your team, but it's also like building that team of experts who can come together to focus on the mission and really deliver.
If you don't have, I'm very good at selling a clinician on Viz. This is obvious to me. And I think when I talk to a clinician, it's obvious to them. But then there's all this process of getting something through a hospital that involves paperwork and moving things through. And that takes a different mindset to me. It takes a bird dog sales person who's just going to make sure every step is done diligently. So you've got to have that team around you to succeed.
And you've also got to focus on the burning challenge or the biggest hurdle. You can't focus on everything and not everything needs to get perfect. Companies, successful companies, usually have only got a few things right along the way. And you look at some of the things they do and you're like, oh my God, how did they succeed there? Quality management system's terrible or something like that. And actually the reality is you only have to get certain things right. But if you don't get those right, you die. And so focus, so growth mindset, build an excellent team and then really focus on what that imminent challenge is that you have to overcome no matter what to get to the next stage whether it's FDA clearance or reimbursement or learning how to sell whatever it might be focus on that and don't worry about almost everything else.
Dr. Tim Showalter: Just to build on that, I think a lot of folks who listen to this podcast are young physicians or just at the start of their career and who have that entrepreneurial bug in them like you had back in 2016 when you came to Stanford. What's maybe one big piece of advice that you would give that you've learned from your time at Viz?
Chris Mansi: The hardest piece of advice to follow, which is enjoy the journey. It is so stressful and so hard. But what's really funny is the things that you found stressful a year ago don't bother you at all. For example, our first FDA clearance, oh my God, that was so stressful. A year later, we were doing four in parallel and none of them were stressful. But we knew what we were doing because you'll learn. And so just knowing that, knowing that your future self will find your current thing that's stressful reasonably easy because you'll get over it is really nice to know. And so enjoy it. Enjoy that journey along the way because it's pretty cool trying to build a team and do something no one else has done before but you can get lost in just how hard it is and how stressful it is.
Dr. Tim Showalter: That's great advice. Thanks so much for joining us today. It's been a great conversation.
Chris Mansi: Thank you. Thanks so much for having me.
Dr. Trevor Royce: So that'll do it for this episode of Health Tech Remedy. Don't forget to subscribe, rate, share the show. Subscribers are so important to supporting our mission. We're glad you're joining us. Thanks so much Chris. We'll see you next time.
