Lung nodules are one of the most common "incidentalomas" found in modern imaging, yet the path from detection to diagnosis is often fraught with patient anxiety and unnecessary biopsies. In this episode, we are joined by Chris Wood, the CEO of RevealDX, to discuss how his company is revolutionizing the diagnostic process. By using advanced AI to move beyond simple detection and into precise characterization, the platform helps clinicians distinguish between benign spots and high-risk malignancies before invasive procedures are required.
Chris shares his journey as a serial entrepreneur and details the rigorous process of building clinical evidence for AI-driven tools. We explore the critical role of radiomics in identifying subtle features invisible to the human eye and the complex landscape of health tech reimbursement. The conversation also covers how these tools can streamline hospital workflows, support overburdened radiologists, and ultimately improve the success rate of early cancer detection programs by catching malignancies when they are most treatable.
If you are a clinician, hospital administrator, or health tech founder looking to understand how to bridge the gap between innovative algorithms and routine clinical practice, this episode is essential listening. Chris provides a masterclass in navigating the "second wave" of healthcare technology, focusing on sustainable business models and measurable patient outcomes. Learn how a risk-adapted approach to lung nodules can save lives while reducing the burden on an overstretched healthcare system.
Episode Resources
Introduction to Health Tech Remedy
Dr. Trevor Royce: Hey.
Dr. Tim Showalter: Good morning. Are we live? I think you and i are live at least. I'm not sure if Paul's caught up with the real moment yet.
Dr. Paul Gerrard: I think I am.
Dr. Tim Showalter: You look alive. Okay then, I think you're with it. I'm jealous you guys get to work together now.
Dr. Trevor Royce: It's fun. I gotta tell you, life on the airplane is not always fun. I had the worst flight of my entire life last week. I was going to visit Paul, but I spent the entire flight in the bathroom heaving with norovirus.
Dr. Tim Showalter: Oh, because you had the GI plague in your house.
Dr. Trevor Royce: Yeah, it hit me during takeoff and the flight attendant would not let me in the bathroom. So I just sat there at the seat. I feel really bad for my seatmate.
Dr. Tim Showalter: Did you have to use one of those snack bags, those little roll-up bags?
Dr. Trevor Royce: I used the behind-the-seat bag and then I never got to see Paul because I quarantined myself in my hotel room for the next 48 hours.
Dr. Paul Gerrard: And we're all happy for it.
Dr. Trevor Royce: And you look great, Paul. You look great. I prefer you virtually, actually.
Dr. Tim Showalter: You do look svelte, Trevor. You look at the Hollywood diet.
Dr. Trevor Royce: Exactly.
Dr. Tim Showalter: All right, should we hop in?
Dr. Paul Gerrard: Well, I don't think we're going to give anybody GI plague, but it might make them feel like it.
Dr. Trevor Royce: Welcome to Health Tech Remedy, the show where three physician leaders in health technology break down the stories behind today's most innovative health companies and speak with the leaders shaping the future of care.
I'm Trevor Royce, radiation oncologist and researcher with experience in real-world evidence, informatics, and AI diagnostics.
Dr. Paul Gerrard: And I'm Paul Gerrard. I started my career in physical medicine and rehabilitation before focusing on reimbursement policy, molecular diagnostics, and market access for AI-driven products.
Dr. Tim Showalter: And I'm Tim Showalter, a radiation oncologist and former med device entrepreneur, now focused on scaling AI technologies that improve care for cancer patients.
The Clinical Problem of Lung Nodules
Dr. Trevor Royce: This week, we're diving into RevealDX, a company applying AI and radiomics to one of the most common and most consequential problems in diagnostic imaging, the identification of lung nodules.
Dr. Paul Gerrard: RevealDX is focused not just on finding nodules, but on helping clinicians decide which ones actually matter.
Dr. Tim Showalter: And I think it's that distinction—the detection versus characterization—where things get really interesting. Maybe we can start by focusing a little bit on the problem overall.
Dr. Trevor Royce: Fundamentally, what's the problem that RevealDX is solving for? I feel like this is probably underappreciated amongst the lay public, but as clinicians, we're during our training, it's beaten into us to become acutely aware of the costs of imaging.
What I mean by costs is not necessarily financial costs, but the downstream consequences. If you get a CT scan, inevitably you're going to find something. Often that's lung nodules or little spots in the lungs that may or may not be a problem.
This can cause issues like anxiety for the patient that you've now found something that previously you had no idea was there and actually may never bother you at all. There is a direct consequence of identifying this new nodule.
But of course, once you find it, you have to do something about it. You've got your CT scan, you see something in the lung, and what happens now? That's fundamentally what these guys are trying to help out with. This has basically exploded with the introduction of three-dimensional imaging in healthcare for so many different reasons.
Dr. Tim Showalter: Trevor, there's the incidentaloma term. We have a medical term for these things that are found on CT scans. When you think about the problem at scale, like if you're managing a hospital radiology department, the number of these nodules that are going to be identified in particular is pretty remarkable.
If you focus on the health system level, this becomes a huge challenge to make sure that patients don't fall through the cracks and that you've prioritized things in the right way. It's important to have some support for that overall.
I think the critical distinction here is that RevealDX is really focusing on applying radiomics to characterize these nodules and do a better job of providing support for identifying which of these are really high risk and may actually represent cancerous nodules versus things that are benign. That's the support that people need.
Dr. Paul Gerrard: I feel like every second CT scan I had in a patient over the age of 65, there was a lung nodule. A radiologist gives a Fleischner a Lung-RADS score and they all needed a follow-up CT scan in three or six months.
You're trying to make sure their primary care provider knows about it and gets it done. The impression I've gotten is that those systems are intentionally conservative to avoid missed cancers, which I think is appropriate.
But then you end up with a lot of scans. If you're the patient and you've got a lung nodule and somebody says this could be cancer, and oh, by the way, we're going to wait three or six months to make that decision—that doesn't exactly put you in a happy place for the next three or six months.
Dr. Trevor Royce: This is a great term, incidentaloma, Tim. I'm glad you brought that up. That can apply to any image that you get, like a CT scan of the abdomen or an MRI of the brain. Anytime you're looking at something, you might find something that you weren't initially looking for, hence your incidentaloma.
Tim, I don't know if you want to reflect a little bit on why the lungs particularly have outsized relevance in this concept?
Dr. Tim Showalter: The lungs are particularly compelling because it's really the challenge that reflects the system at scale for the most part. One way to think about why this is such an impact is that there are compelling data for using low-dose CT scans as screening to identify patients who are at risk for lung cancer.
There are a lot of people in the U.S. who should be receiving low-dose Chest CT screening to identify lung cancer. Then there are plenty of patients who show up in an emergency department with some sort of acute condition and would have a Chest CT.
It would be the standard part of workup for many conditions. All of the other sorts of scans that are done being ordered for other purposes can identify these small lung nodules.
Dr. Trevor Royce: I think that's right. There have been these large population studies that have shown a benefit to doing screening, basically giving people that have the right history a CT of the chest because they've been smokers or whatever.
We walk around the world and breathe dirty air and all of us are going to have certain little spots in our lungs. It becomes a question of whether that spot is meaningful or not.
Reveal AI Lung and Risk Stratification
Dr. Tim Showalter: Yeah, and I think this is where maybe it's good to talk about RevealDX's core product. It's called Reveal AI Lung. It's an interesting tool that's focused on characterization, not just detection.
One of the challenges is that in a radiology report, you might have a description of like, "we see a nodule, it's a certain size, it's there." And then the rest of the care team is left to follow up on that.
RevealDX's product focuses on an index score, or a malignancy similarity index (MSI), that estimates how similar a given nodule is to known malignant nodules. By providing a score to it, it provides additional support for triaging or framing the patient's individual risk.
As Paul pointed out earlier, a lot of the frameworks that have been otherwise provided before this are pretty conservative tools. It's helpful to have more of a stratification of risk.
Dr. Paul Gerrard: If you're a clinician now, you're looking at something that's not just a binary call—hey, at risk of malignancy or not—it's telling you we think it looks like malignancy, here's how much, and then that helps you.
You always get that "nodules of unknown significance, clinically correlate." Well, now you have something to clinically correlate with because you can say, "Oh, this is a high risk. I already had a high pretest probability. I'm at a particularly high index of suspicion for this patient having cancer." This brings that index even higher.
Dr. Tim Showalter: There's still the clinician in the loop. That's what a lot of people want to see from care. There's still a physician making a decision or a care team making a decision about how to approach these. Having that score provides an additional level of support.
Building Clinical Evidence
Dr. Trevor Royce: A lot of the efforts today have been building the evidence and showing how their product performs. That means writing manuscripts, submitting to journals, and having it reviewed by peers.
They've had some success there where they show how their score performs in terms of risk discrimination, how much statistically does it reduce things like false positives—meaning that you have a dangerous lung nodule, but actually it isn't—and comparing that against the standard of not using a tool like this and just having the radiologist involved.
Dr. Tim Showalter: Overall, what is interesting about laying this on top of just simple size measurements is that you can get some movement on reductions in false positives and population-based metrics that can help a system handle a large throughput for patients.
They've got some peer-reviewed reports that have added evidence behind that. I expect to see more from them over time.
Dr. Trevor Royce: I think some of the numbers that we saw were around a 30% reduction in false positives compared to baseline workflows. It may be worthwhile just to quickly play out what that means by a false positive.
That's the idea of saying, "Oh, we found a nodule. We think it's bad. We need to do something about it." Often the next step there would be to stick a needle in it and do a biopsy. There can be downstream consequences to that.
You could injure a patient by giving them an infection or they could have their lung collapse after the biopsy and they become hospitalized. These patients have bad lungs because they have a history of smoking. All of a sudden you have a patient and a couple nights in the hospital for a nodule that actually was nothing. That's what we mean by false positives and the cost of this.
Dr. Paul Gerrard: They had a COPD and now they're on a vent because of that lung biopsy.
Dr. Trevor Royce: Exactly what you want to avoid.
Dr. Tim Showalter: That's a rare case, but it's not a made-up story. There can be real harm.
Dr. Trevor Royce: These risks are low on the individual level, but when you apply these approaches across populations, the numbers add up. You start having large numbers involved.
The Importance of Early Detection and Payer Alignment
Dr. Paul Gerrard: That's really critical here. Right now there's a lot of discussion in the world about catching cancer early, and that is important, but there's the flip side of it. Okay, you've caught something suspicious for cancer. A lot of those are going to be non-cancerous.
In a world where we just had a bill passed that would cover multi-cancer early detection blood tests, which I think is great, you still got to figure out how we are going to reduce those false positives in places where we already have screening programs set up.
I think being able to rule out these cancers is in some respects a little bit of a movement in the opposite direction that we see happening in society. But it's a move where we already have established screening programs set up. This is going to become even more important as cancer screening becomes more prevalent in more cancers.
Dr. Tim Showalter: I've read that RevealDX has clearance in Australia and New Zealand. Obviously, there is a stringent evidence bar there. It certainly differs from FDA standards. My understanding is they're working towards additional validation for U.S. FDA clearance as well.
Dr. Trevor Royce: And Tim, I think you've been doing a lot of work in this space. When you get these approvals, then you can start collecting data in other ways, either through real-world use or prospective data collection. It can be a positive feedback loop.
Dr. Tim Showalter: Well, I think they do have some reports out there for some interest for U.S. health systems. It'll be interesting to follow how that plays out over time.
A lot of health systems are thinking about how they use technology to manage their lung nodule workflows and support their teams that are managing that. There is some interest in the U.S. This is a space that I've had some involvement in recently.
It's a landscape where there are a lot of tools available. None of them are absolutely perfect. But I think it's part of an overall strategy. Something like this where you have a stratification could be really helpful for an overburdened health system trying to do the best job to serve patients.
Dr. Trevor Royce: Yeah, and I think in a crowded space—and Paul, I'd love to hear your thoughts on this—an important piece of demonstrating the value of a tool like this is showing how it actually changes clinical management and changes decision-making. How does it influence what the providers ultimately do and how the patients actually do?
Dr. Paul Gerrard: That's obviously going to be the hardest part—how do the patients ultimately do? Especially if you're detecting early lung nodules, getting the outcomes on that can be a while.
Ideally, what we will see is that this will reduce unnecessary biopsies and unnecessary imaging, and that patients are going to do just as well 10 years down the line, if not better.
Dr. Tim Showalter: What payers think matters a lot and reimbursement is going to be key for deployment of solutions like this. Obviously, we're all in the AI space. Can you give a little bit of comment on what they need to show to payers? How critical are reimbursement considerations in this?
Dr. Paul Gerrard: Yeah, well, I think one of the things to mention here is that there can be a lot of politics in this too. If they were to pick up support of some kind of major guideline like USPSTF or something, that has its own pathway and its own implications.
If you're just saying what would payers necessarily want to see within the evidence, one of the big things is they really want to understand what is the ultimate outcome on patients. That is what they want the plausible story on.
Ideally, they would have that evidence that looks five or 10 years out that shows that there is reduced utilization with preserved outcomes or similar utilization of care with improved outcomes. What they are really worried about is that they would cover a new technology that would potentially worsen outcomes, even if it reduced utilization.
Contrary to popular belief, obviously payers do want to save money, but you don't want to be the payer that's covering a novel technology and worsening patient outcomes while doing it.
The other thing is if you say, "Okay, we're going to get this," what's somebody going to do after they have the results? This is where guidelines will become important because if guidelines are still based around things like Fleischner or Lung-RADS, you'll end up having doctors get the CT scan and say, "Okay, I need to re-CT them in three or six months. I need to biopsy them."
Sure, this RevealDX score does give me a little more information, but if I'm still following a guideline-driven care pathway and the guideline has not yet incorporated the technology, it's not going to change my management at all. Payers don't like to see that either.
Payers want to pay for science that is implemented in practice, not science that is theoretical and could be implemented in practice. Until the field catches up with the science, it's not necessarily changing the patient's course of management.
Dr. Tim Showalter: That makes a ton of sense. I'll also add like as a simple rule, if the evidence supports payers covering this and payers can generate revenue for using this, it's the way to get this technology implemented. If there is some sort of an opportunity there, I think it would really be a way to drive implementation overall if it does truly move the needle for patient care.
Dr. Trevor Royce: One takeaway from what Paul was talking about is that the adoption of these types of technologies don't happen overnight. There's a lot that needs to happen before it becomes widely adopted.
That's probably the system working as it should when you're dealing with human lives. Just having the evidence and building this cool algorithm that detects nodules isn't good enough. That's just part of the story.
You still have to get into guidelines, show that it's useful, and show what the impacts on patient outcomes are and all these downstream things. That takes a lot of time and energy.
Guest Interview: Chris Wood, CEO of RevealDX
Dr. Tim Showalter: Yeah. I think that our listeners are going to want to stick around. We are fortunate to be joined by the CEO of RevealDX, Chris Wood.
It's a really interesting conversation that we'll have focusing a bit more on his history as a prior founder in a pretty similar space overall for leveraging technology for cancer detection and hearing more about what he's up to with RevealDX. Please stick around for the interview with Chris Wood.
Dr. Trevor Royce: Today, we're thrilled to be joined by Chris Wood, CEO of RevealDX. Chris is a serial entrepreneur who has founded and built two companies that have successfully exited, including the first company to achieve FDA clearance for an automated cancer detection in breast MR. Welcome to the show, Chris. Excited to have you.
Chris Wood: Thanks for having me. It's nice to be here.
Dr. Trevor Royce: Just to kick things off, we'd love to hear some background on RevealDX, maybe its origin story. Was there an initial insight or moment that led to the pursuit of this? Tell us a little bit about the history of Reveal.
Chris Wood: It's maybe a little bit related to your origin story with this podcast. I sold my last company and had to do my two years at the acquiring company. After that, I knew I wanted to start something else in medical imaging AI, because there's a lot of growth in that industry right now.
I didn't want to spend the next 10 years building a company. So I gathered a bunch of friends that I had known from the industry for quite a while. They said, "Let's put our money together, find something that looks like it has a lot of potential."
We started looking around for companies. We actually bought a company during COVID that was having a hard time raising capital, mostly because of that and a couple of other things. I stepped in as CEO.
We found it really super attractive because it was solving a big problem. Also, there was a possibility that it would be reimbursed, which is the holy grail of medical imaging AI. There are a few companies that have achieved this, including HeartFlow and a couple of others, but it's not common. So we decided to take a run at it and it's been five years. We're almost there, we hope.
Professional Background and Educational Origins
Dr. Tim Showalter: Great, Chris. I think it's a fascinating story and I'm curious to go even a little farther back. Can you give our listeners a little bit of maybe an origin story for Chris Wood? What's your professional background and educational background before all this?
Chris Wood: Like a lot of kids that were my age, I wanted to be an astronaut or a nuclear physicist because science was really hot back then. I ended up actually studying physics in school.
In graduate school, I had the opportunity to either do laser systems to build guidance for missiles—that was one funded project—or medical imaging was the other one. I decided to go with medical imaging. I worked at Moffitt Cancer Center in graduate school and got a degree in medical physics and then went right into industry.
I never practiced clinically, but that was the beginning. When I was in grad school at Moffitt, I remember the first time I wrote software to display some of these images, and I just fell in love with it immediately. I knew it's exactly what I wanted to do with my career, and I've stayed in medical imaging software since then.
Right around that era is when we were replacing film-based reading with digital reading. We had all this opportunity with 3D acquisitions to do really cool visualizations of medical imaging data. We also had the reduction of film and replacement with PACS. So there was a ton of software to write. It's kind of kept me busy for the last 35 years—being involved in the whole transformation of radiology to the digital world that it's in now.
Dr. Trevor Royce: Amazing. Something tells me young Chris Wood and young Trevor Royce would have been great friends. And I suspect Tim Showalter as well. We're both radiation oncologists, and radiation biology and radiation physics play a large part of our professional world.
Chris Wood: Yeah. And doing grad school at Moffitt was an amazing experience. I was right there in the nuclear medicine department and we were dealing with patients. We were writing software and that software was being used on patients.
To see that impact was really profound for me and just hooked me to this whole industry. I never looked back after that point.
The Crucial Role of Reimbursement in AI Adoption
Dr. Trevor Royce: I want to quickly go back to two comments you made about the genesis of RevealDX. There were two things: one was a solvable big problem and two, the reimbursement piece. Can you give the listeners a little bit of context on that reimbursement piece? Because I think that's so important to unlocking how this can actually reach patients.
Chris Wood: AI has moved a lot faster than Medicare and the payers. A lot of AI products have actually made it through FDA pretty easily, relatively. So there's hundreds of medical imaging AI products out there.
Almost all of them need to be supported through the radiologist actually taking money out of their pocket or the health system taking money out of their finances and paying for them because they feel like it's worth it, not because they get reimbursed by insurance, even if it has a positive impact on outcomes.
Ten years ago, mammography CAD was like the only thing reimbursed. But now we've got pathways for reimbursement. You mentioned HeartFlow; they started out with a Category 3 code, which was reimbursed. Then they finally converted to a Cat 1 code, which means it's in the Medicare physician fee schedule.
Clearly, another company that has gotten plaque characterization now is going to be converted to a Cat 1 code at the first of the year. So we've got these kind of established pathways now. Reveal, our software, you can bill it against a Cat 3 code currently.
It's part of our responsibility as the vendor, as well as others that are in this space, to continue to add that clinical validation to drive up usage of the software and to do the health economics analysis to convince the powers that be that this thing should continue to get paid.
We feel like we're in a very, very good spot because anything you can do to accelerate diagnosis of lung cancer is really going to have a big impact on patients as well as cost. We feel like we've got a really clear path to getting this thing converted to Cat 1. We feel very confident about that.
Characterizing Nodules Through Radiomic Biomarkers
Dr. Tim Showalter: From my understanding, we've gone a step further with RevealDX in that in addition to identifying lung nodules, your product does some work to actually characterize the lung nodule and define behavior. Can you go into a little bit more detail there?
Chris Wood: We've got this incredibly sensitive test, which is Chest CT, that finds nodules today that are five millimeters, even smaller. Very sensitive test. If you have early-stage lung cancer, it's really likely you're going to be able to see it with Chest CT.
But those cancers are mixed in with a whole bunch of benign nodules. The theory behind RevealDX is that if you feed enough multidimensional information and the right feature set into an AI system, it will be able to see things that radiologists cannot see that characterize that nodule and allow you to pick out the ones that are at higher risk.
Think of it as the nodule kind of expressing radiomic biomarkers. Subtle features that maybe a rad can't appreciate with their human visual system. Maybe they're more three-dimensionally oriented. Maybe they have more to do with their surrounding parenchyma.
The AI system is able to synthesize all that information. We look at thousands of features and then immediately produce a score, which can then inform whether or not this particular nodule is one that needs to be followed closely.
The bigger health systems out there can find 1,000 nodules a month. Tracking all of those patients and making sure that they follow up right on schedule is often very difficult. You can't, from a health economics standpoint, follow all of those super closely.
You end up getting patients with small nodules and tell them to come back in a year. They may or may not come back. I think it's like under 50% will even come back at all.
That doesn't necessarily present itself as a big problem right away because the vast majority of those small nodules are benign. But if you look at the National Lung Screening Trial data, about half of all cancers first present as small nodules.
If you can pick out which ones to act on and to be aggressive about your diagnostic process with, you can accelerate diagnosis for a good portion of these. Of course, that translates to huge savings. The data right now that's been coming out from I-ELCAP and others show that the survival rate is fantastic if you can surgically intervene early.
The I-ELCAP data says the 20-year survival rate is over 80% for small cancers that are found on Chest CT that are acted upon surgically. It's a curative process. We want to put more patients into that category as quickly as we can and not let them progress to that late stage where 70% of all cancers are currently diagnosed.
Once you're in that late stage, your five-year survival rate is terrible. We want to catch them as early as possible, but not bankrupt the entire health system while doing it. You guys know you can't stick a needle in everything you find in a lung and find out through pathology whether it's cancer. The complication rate is really high for biopsying lung tissue.
I saw it as this perfect storm. This is a problem that needs to be solved without intervention, if possible. There's just this opportunity to improve it so much compared to what's happening today.
Dr. Tim Showalter: Great.
Dr. Trevor Royce: Yeah, that's really well presented. Lung cancer is the most deadly diagnosis in the U.S. and contributes to the most cancer deaths. It's preventable a lot of the times. That's why things like lung cancer screening are supported by guidelines.
My practice is at the VA and they have a very robust lung cancer screening program, but one can see it being overwhelmed by all these images coming through and the need for radiologists to know what to do about all these lung nodules. Using AI to identify these makes a lot of sense for solving such a huge problem that can help so many patients.
Optimizing Radiologist Workflow and Impact
Chris Wood: The joke right now in radiology is that when you enter the emergency department, you might as well just walk through the CT scanner, because so many CTs are done on people that come into the ED currently.
CT scans are just getting so ubiquitous. We do 20 million a year in the United States, which means we're finding about 8 million lung nodules with the resolution they're getting now. We're finding more than 20,000 nodules a day. That's just in the U.S.
You can't say, "Okay, we're going to follow each one of these patients very closely for the next two years until we find out what happens to this nodule." If there's a way—and we feel like we've got a way—to use artificial intelligence to look at that data closely and pick out those nodules that we really need to be concerned about, that will do nothing but help the health system and patients.
All of this stuff isn't completely figured out yet. We work with individual providers to figure out exactly how they want to deploy this technology. But we do have a tool that we know works. We've validated on over half a dozen studies all around the world.
Every time, we're able to outperform the diagnostic process standard of care today by adding this artificial intelligence in. In the aggregate, we can detect about 45% of all lung cancers faster than you would using standard follow-up techniques according to the guidelines. You've got potentially half your cancers where you can accelerate diagnosis. That's going to really move the needle.
Dr. Trevor Royce: To that end, how has it been received by the healthcare providers or the radiology community? Culturally, people always think of AI as sort of existential—there are all these challenges of how it'll be used. Can you reflect on that?
Chris Wood: Radiologists do not want to be slowed down. That's written in a tablet somewhere. The reason for that is because they're completely overwhelmed and overworked. It's tough to say to them, "Here's something that can potentially improve patient care, but it's going to slow you down a lot."
We had to do a lot of work to make sure that this was super fast and super easy. But the response has actually been quite good. One reason is the reimbursement that this particular test happens to have. And the next is just Chest CT in general.
If you look at all the medical imaging exams that are performed, Chest CT is the slowest one to read. Looking at lung nodules is a small part of reading a Chest CT. It'll take on average between 9 and 12 minutes to read a Chest CT. That's slow for a rad.
There's a lot you need to look at there. It covers a lot of the anatomy, the heart and lungs and everything else. You're taking a look at everything where things go wrong with people.
A little bit of help with this particular problem, which also happens to have big implications for the health of the patient but also medical legal issues related to this—you know, a small nodule with a very low likelihood of malignancy, that patient could come back a year later and have stage four lung cancer. Then you've got an issue.
This is a tool that allows you to say, "Look, we're dealing with these nodules using the best technology we have." It just makes reading that Chest CT a little bit easier. We don't claim we speed them up, but we also don't slow them down.
We provide a more valuable report that benefits the radiologist because they want to add as much value as possible and also provides this additional boost of reimbursement.
We're trying to get professional fees as well. Right now we have a technical fee. When we get that professional fee, we'll actually be able to reward the radiologist for all the time they spend looking at the score that we provide and interpreting it.
Right now they cover that cost, but at least the technical fee is there to benefit whoever owns the CT scanner and to try to fund these programs to manage nodules more effectively. That's almost universally what we hear from customers.
They're not looking at this to save their hospital from its economic woes. It's not enough money to do that. But there is enough money from this code to do things like hire nurse navigators, buy software they need to manage nodules, and improve their lung cancer management programs. All of them want to do that. They're all understaffed in that regard.
The Hospital Perspective and System-Wide Benefits
Dr. Tim Showalter: Chris, you started off early in the conversation referencing just how many scans are done. It seems like every hospital needs to have a lung cancer screening program or small nodule clinic.
What's been the reception of the hospital administrators and the health system leaders overall? Have you found that they're the stakeholder that owns the scanners? What's the reception been to your technology overall?
Chris Wood: Certainly, I think there's two or three different positives I can point out. One is some of them are interested in it from a marketing standpoint. AI is sexy right now. If they're trying to build lung cancer screening in their community, they can say they've got this AI-enabled technology at their hospital, and that can be attractive.
The other is obviously the reimbursement. We've talked to health system administrators who didn't flat out have the money to invest in more of the nodule management infrastructure that they wanted to, and they look at this as a way to fund that.
We're not taking the entire reimbursement as a company. We're revenue sharing that with the health system. It can help cover some of the expenses of things they really wanted to do.
The third is that there's a lot of competition for doctors right now. Thoracic surgeons, pulmonologists, radiologists—they all have a lot to do with this problem. They're all interested in this technology. If you as a health system administrator can keep your thoracic surgeon happy, you're going to try to do that.
If you're a thoracic surgeon who just started using your robotic bronchoscopy system, the goal of that system is to try to be able to biopsy small nodules. Eight out of 10 of these biopsies turn out to be benign. You're not going to feel good about using your system that way.
But if you can find something that'll identify higher-risk nodules in your program, then you're in there biopsying malignant small nodules a lot. You're having this positive impact on the patient with this new tech. That robotic bronchoscopy technology is actually accelerating interest in our product.
Managing Challenges and Educating Users
Dr. Tim Showalter: That's great. You've really hit on a spot that serves a lot of needs and ultimately benefits the patients because you're providing a risk-adapted approach for them. What's been the most surprising challenge along the way for developing this? What was the bottleneck maybe that you did not anticipate?
Chris Wood: I think one of the things that always surprises me is that there's some education that needs to be done when using a product like ours. There's a tendency for us as humans to trust that AI is just going to give us the answer.
You go to Claude or some tool out there and it's really upsetting when you don't get the right answer because we're thinking that this is kind of a magic technology. One of the things we need to educate people about is that we're showing through our score that the nodule is expressing certain biomarkers and it's a higher risk. This is not a virtual biopsy.
It would be great if it was. It would be great if we could just run the tech on a nodule and say, "Yep, that one's malignant, 100%. This one's benign, 100%." It's not there yet. We need to educate people that that's not the way AI works.
If you integrate this into the program, you can really move the needle. You can really improve early detection within your health system, but it's not a magic bullet. I've been working in AI so long that I didn't expect some people would feel that way, but we have had that happen.
There's an education process of, "Well, if it's not 100%, then how do we actually use this thing?" As a company, part of what we need to do is provide support for people who adopt this technology to help them understand how they can use it properly in their specific health system.
There are variations in health systems—what their goals are and what their particular problems are. You really need to partner with a company when you're deploying technology like this. It's not just "pick it off the shelf and it gives you this 100% answer." It is a partnership.
As we get more and more customers, that'll get easier and easier, as it did in Europe. We've seen that happening.
Workflow Integration and Delivery of Results
Dr. Trevor Royce: Following up on that from the provider perspective, this tool can make the healthcare system more efficient and help these providers be more efficient. It's a very useful tool in our toolbox.
For our listeners, there are all sorts of concerns about physician shortages. This is a great example that can help mitigate some of the specialty shortages. Beyond that, can you tell us a little bit more on the product side? Like, how this is integrated in clinic and what it looks like from the physician perspective? How are the results delivered?
Chris Wood: We decided that the place to deploy our technology was at the point at which the nodule is found. That means inserting it into the radiologist's workflow.
We decided not to deploy this technology on nodules after they're referred to a nodule clinic or tumor board, because our tech works really well on small nodules and we wanted to find early cancers.
The way our software is deployed is that it's deployed at the point at which the radiologist finds the nodule. That can mean a single click by the radiologist, which then sends those coordinates to us. We compute the score and we send the score back. That's right inside their PACS.
If the radiology group uses something called a nodule detector, then the coordinates for the nodule can be sent to us a priori before the radiologist even sees the scan. Then our score—which we call the malignancy similarity index—is presented right next to the nodule when the rad pulls it up.
So it's super easy workflow for them. The easiest way to go to market really is by partnering with nodule detectors and PACS so that we can go right into their specific IT environment, integrate with them, and minimize the impact for the hospital IT department. We're kind of bolted onto the PACS.
Looking Ahead: 2026 and Beyond
Dr. Tim Showalter: Great. Curious for you, what's next? What are 2026 and 2027 looking like for RevealDX?
Chris Wood: I mentioned that we are in the market and currently installed clinically in Europe at a number of sites. We are only installed in the United States under IRB because we are in the process of getting through the FDA.
I risk jinxing myself if I say we plan to get through by the end of the year, but that's our plan. Right now, we're pretty confident we're going to be able to get through the FDA because we did submit in mid-2025. We're hoping to get that final decision.
In 2026, we'll be scaling the company in the United States, continuing to build our market in the EU, and doing more partnerships. The PACS industry is pretty fragmented. You can get 20% of the market by going with the leading PACS vendor, but you're not going to get 80%.
The last company I started, we had integrated with over 30 PACS. We'll be doing more integrations as we bring on more customers. Then, of course, we've got other ideas looking up and down the diagnostic process in lung cancer. There are a lot of problems to be solved there and AI can help with some of those.
We don't currently have plans to break out into other cancers right now, but we'll see if it looks like our technology can really help another cancer. But right now, we're a lung cancer company.
Dr. Trevor Royce: Amazing. Well, we're certainly excited to see how the next five years unfold for you all. We'll be following along closely.
Chris Wood: Well, thanks a lot.
Final Thoughts and Advice for Health Tech Entrepreneurs
Dr. Trevor Royce: As we come to our close here, final question for you: as you reflect on your career and journey, do you have any advice for entrepreneurs or clinicians as they consider their own adventures in health technology and starting their own AI companies?
Chris Wood: I think it's a good time to come into this space. It's sometimes a little difficult to be the first mover. We saw a lot of AI companies come into being in medical imaging right at the time when people were saying radiologists are going to go away and this is going to be completely disrupted.
A lot of those companies are really struggling. I look at what's happening now as the second wave, where we've got real business models behind these companies. We've got reimbursement, which is going to create revenue opportunities that are going to fund the entire medical imaging AI industry.
If you're going to build a healthcare AI company, the Silicon Valley approach of "move fast and break things"—that's not really the one that generally works best. It's best to be thoughtful, maybe a little patient, and think about the economics, the entire ecosystem, and all the physicians who are affected by this type of change to their work.
If you can establish those partnership-type relationships with the providers and you can work with them to solve their problems, they become incredibly loyal customers. If you get to the point where you understand them and you can get them to trust you, then you can be a successful entrepreneur.
If you're looking to build a rocket ship that's going to have a $12 billion valuation six months after you start it, go into something else. Don't go into healthcare. But if you're thoughtful, maybe tackle things a little more scientifically, have a little bit of patience, and you like partnering with customers, then maybe this is a good field for you. I feel like it has been for me.
Dr. Tim Showalter: That's fantastic. Chris, we should close there. That's a great summary. It does inspire the future generation of entrepreneurs by highlighting the virtuous cycle of healthcare—you get strong science and you get to make an impact. Thanks so much for joining us. It's been amazing to hear about your journey with RevealDX.
Chris Wood: It was fun. Thanks for having me on.
Dr. Tim Showalter: That's it for this episode of Health Tech Remedy. Don't forget to subscribe, rate, and share the show. Subscribers are so important to supporting our mission. See you next time.
