The Future of Lung Cancer Detection: Chris Wood & RevealDx’s Revolutionary Approach
Episode Topic: In this episode of Skeleton Crew, we embark on a journey through the cutting-edge radiology landscape, focusing mostly on groundbreaking advancements in lung cancer detection. Our guest Chris Wood, Chief Executive Officer of RevealDx, enlightens us about the transformative technologies and methodologies employed by RevealDx, aimed at significantly improving lung cancer detection and subsequent patient care.
Lessons You’ll Learn: Tune in to gain profound insights into the rapidly evolving field of radiology, the pivotal significance of early lung cancer detection, the game-changing role of Artificial Intelligence (AI) and Radiomics in revolutionizing diagnostic accuracy, and the compassionate approach of non-invasive methodologies toward patient-centered care. Get to know how RevealDX’s strategic collaborates with Sirona Medical’s partnership with RevealDx and revolutionizes lung cancer screening through cutting-edge AI tools.
About Our Guests: Chris Wood, our esteemed guest, is a seasoned entrepreneur and accomplished medical physicist with a remarkable career trajectory. Currently, at the front line of RevealDx, Chris is spearheading a mission to redefine lung cancer detection through state-of-the-art technologies. Chris’s exceptional track record of securing FDA clearances for automated cancer detection underpins his expertise and unwavering commitment to advancing healthcare paradigms.
Topics Covered: During this engaging conversation, we traverse Chris’ professional journey from a medical physicist to the helm of RevealDx as CEO. We delve into RevealDx’s innovative approach to lung cancer detection, placing a spotlight on the pivotal role of Radiomics and AI. Moreover, we contemplate how RevealDx envisions shaping the future landscape of radiology and healthcare, aligned with patient-centric principles and geared toward enhancing overall patient outcomes.
Our Guest: Chris Wood, CEO of RevealDx
Meet Chris Wood, a pioneering force in the fight against cancer through early diagnosis using medical imaging. He is the Chief Executive Officer of RevealDx, a company dedicated to transforming lung cancer detection. With a strong background as a medical physicist and extensive experience as a CEO and CTO, Chris has an exceptional understanding of the radiology industry. He has also previously founded successful startups in medical imaging software, both of which were acquired by larger companies.
Chris firmly believes that empowering radiologists with advanced tools is the key to delivering more clinical value, ultimately contributing to a healthier and happier world. At RevealDx, he leads a team that’s at the forefront of revolutionizing the detection of lung cancer. Their breakthrough technology allows for early and non-invasive identification of risky lung nodules, a crucial factor in cancer cure rates.
Through his journey and dedication, Chris has shown that innovation and a patient-centered approach can redefine the landscape of radiology, significantly impacting healthcare.
Follow RevealDx on
Chris Woods: What I’m trying to do right now is take some of that good work and bring it to the commercial market, because I’ve had the experience of taking some good CAD type work and seeing the impact it can have on patients. It’s pretty profound. I mean, it’s one thing to do something in a lab and then maybe even deploy it in your own clinic, but when you can get a commercial product out there with wide adoption, the impact you can have on patients is orders of magnitude more.
Jennifer Callahan: Welcome to the Skeleton Crew. I’m your host, Jennifer Callahan, a technologist with ten plus years experience. In each episode, we will explore the fast paced, ever changing suburbs. Completely crazy field of radiology. We will speak to technologists from all different modalities about their careers and education. The educators and leaders who are shaping the field today and the business executives whose innovations are paving the future of radiology. This episode is brought to you by X-ray technician Schools.com. If you’re considering a career in X-ray, visit X-ray technician Schools.com To explore schools and to get honest information on career paths, salaries and degree options. Hey, everybody. Welcome back to another episode of The Skeleton Crew. Today I have a true visionary with me. His name is Chris Wood. He is the chief executive officer of RevealDx. And it’s an innovative company in the realm of AI that’s assisting in diagnosing of lung cancer. So it’s really doing some groundbreaking work. So, Chris, thanks for taking the time to be with me today.
Chris Woods: Oh, great to be here. Thanks for having me.
Jennifer Callahan: My pleasure. So first, let’s put the spotlight on you a little bit. You’re doing some awesome work here with the lung cancer detection, but how did you get into this role?
Chris Woods: Going way back, I chose to study physics in school. I think like a lot of kids my age wanted to be an astronaut. That seemed like one of the ways to get there, but I actually ended up really liking the science. And then in grad school, I migrated over to medical physics and just fell in love with it. The first time I was able to display images on the computer screen and start messing around with them and I worked at a cancer center and the clinicians and the technologists and everybody who was treating patients would give me problems to solve, and I was able to problem solve right there in real time with software. And I just fell in love with it and then worked at a couple of big companies. One you might remember is Picker. Another one is Siemens, where I ran software development for the ultrasound group. But every time I was in one of those big companies, I was always trying to start little things that they found annoying, like a new division or a new project, and finally found out that realized that I really was more of a startup person. I liked to be in the early stage and development of an idea and commercialization, so I was able to do that lucky enough to find some partners and have started three companies now, all in medical imaging software. Two of them were sold. We built them to a certain size and sold them. One of them I sold to merge Healthcare and it was called Confirm, and another one sold to Intelsat, which was called Clarreo, which was a radiology workflow software. And now this one is RevealDx and as you mentioned, reveal is focused on lung cancer nodule triage.
Jennifer Callahan: That’s great. So I’ve definitely heard of Picker. That was probably one of the first rooms that I worked on as a technologist.
Chris Woods: I think they eventually got acquired by Philips. There was a false start. One year they went together to RSA and said, We’re merging and then that got called off. And then a number of years later it actually happened. Picker was owned started by Harvey Picker in like 1900, right after the invention of the X-ray where James McGregor can’t remember which one it was. And then his son took over, grew it quite a bit, provided X-ray equipment for World War Two. It’s actually quite an interesting story. And they eventually just got gobbled up by the big guys.
Jennifer Callahan: That’s great, though, starting off working with two great companies like Picker and Siemens, definitely get your feet wet and get the exposure to the different technology from a while ago, but it’s definitely helped you get your mind moving into different directions. And like you said, you’re always trying to come up with a different new project or department to add onto it. But right then when it’s found you annoying. But I’m sure that they definitely found some value in what you had there.
Chris Woods: Yeah, we started a division actually in Picker called Image Guided Surgery, and we worked in Frameless, Stereotaxy, and that led to a lot of other opportunities for Picker down the road. So I don’t think they regretted doing that. And it strengthened our relationship with the Cleveland Clinic, which is where we were prototyping this device. And so, yeah, new ideas to have a place in, in these big companies. But I do think that eventually it makes sense to take what you’ve learned there and kind of branch out and start your own because there’s no experience like working in a in a startup. It’s, you know, trying to change a bit of radiology. And believe it or not, it’s possible as a small company you can actually make a pretty big impact in this field. It’s a relatively small field. So I think we were able to do that twice before and maybe this will be a third time.
Jennifer Callahan: Yeah. So let’s move into real. What got your mind moving into the direction of trying to develop new technology for lung cancer detection?
Chris Woods: Well, the first company I started was called Conferma. And the reason we started that company is because breast cancer was moving from a from a staging exam. So you’re staging cancer using breast because it’s a highly sensitive technique and people were starting to use it to screen high risk women, people who had the BRCA gene or other family, strong family histories were the idea was to put them into an MRI and Mammo screening program, which was pretty groundbreaking in the International Breast Cancer Consortium, started this idea, and then it was endorsed by the American Cancer Society. So instead of having 200,000 breast MRI done every year, you’re looking at 2.5 million. So we knew these exams were really hard to interpret. So we built a company that would create software that would allow the radiologist or mammographer to read these exams very quickly and efficiently. And that worked. Fast forward to a number of years ago and the US Preventative Services Task Force recommends screening for lung cancer for high risk patients. It’s very reminiscent of what happened in breast cancer screening. Right. And in this case, it’s very easy to identify people who are at high risk for lung cancer because they are usually smokers. Okay. So when the US Preventative Services Task Force makes that recommendation for screening, which they made based on the results of the national lung screening trial.
Chris Woods: Because of the Affordable Care Act. That means it’s covered by Medicare. As soon as they make that recommendation for screening, Medicare has to pay. So now we have lung cancer screening in the United States and it’s paid for. So with that, we knew what’s going to be the big problem. One of the big problems that comes when you start screening patients for lung cancer is you find a lot of nodules. You find in about a third of the time, you find a lung nodule, which is just like a little dot. And those nodules are vast. Majority of them are benign, but some are malignant. And right now what people do is they do an interval scan, so they wait six months, a year or three months, depending on the nodule, and they have the patient come back. And the tragic cases are the ones where the patient comes back and they have something like stage three cancer. He doesn’t. Survival for lung cancer is really poor. The five year survival rate is 1.7 out of ten people. Wow. And you compare that to and that’s why it’s the biggest cancer killer. But the interesting and maybe hopeful thing on the horizon is that the cap, which is this big international lung cancer screening organization, they’ve been following patients a long time.
Chris Woods: And if you can catch that cancer at stage one, the 20 year survival rate is 80%. Wow. So it’s you find people using words like cure associated with lung cancer. If you can catch it early. The problem is, if it goes late, there aren’t a lot of targeted therapies. There’s not a lot of precision medicine that really works for lung cancer yet. So your prognosis goes down really quickly. Also, the other thing about lung cancer is that when you’re trying to diagnose it, you have to take sample tissue from the lung, which is non-trivial. The complication rate is 22% for lung interventions when you’re trying to get tissue, which costs $1.2 billion every year in the United States and causes morbidity and even mortality in some cases, because these patients tend to be old, they tend to have COPD, etcetera. So you can’t stick a needle in everything. You’re flooded with these nodules and the health care system. You don’t want to just scan everybody every month forever. So so it’s a tough problem and one that we’re trying to solve with AI.
Jennifer Callahan: And now going back to like doing biopsies and such. I’ve been part of a circle for Groskop’s where they do the biopsies. Do you know, or certain parts of the lobes are harder to reach than others that might maybe they might not be able to do a biopsy on a nodule in a certain area of the lung.
Chris Woods: Absolutely. If you go to any of these multidisciplinary nodule clinics or tumor boards and they look at these nodules, they’re considering all sorts of things when figuring out what to do next. There’s going to be an interventional pulmonologist in there saying, Yeah, I think I can get this thing right. I’ve got this new fancy robot or it’s in a location I think I can get to. Or maybe they’re saying this is going to be really difficult. You’re looking at the patient age. They’re looking at the comorbidities. They’re looking at all these things to try to figure out what is the best course of action for this patient. And yes, the location of it is definitely a big consideration. There was just a study out that last week or so that hit the news that if you take wedge dissections, which is a pretty invasive thing to do to a lung, 15% of the time, they’re just taking out benign tissue. That’s high. That’s more than we’d want. So you’ve got this nasty intervention option that you try to avoid with these patients especially. And there is technology now with robots and bronchoscopy, as you said, that’s getting giving us the ability to go in and precisely sample tissue. But it’s not perfect yet and it’s still difficult and it doesn’t work for all patients.
Jennifer Callahan: And then also, too, you have to look at the. What could happen. Post procedure. I know as a technologist you’re always for x ray. You’re going to do a post-op chest x ray to make sure that nothing has happened to the lung. That. Everything’s looking okay there. So yeah. Mean lots of different things going into it. So looking at early cancer detection, then, what are the main sources besides doing a biopsy? Will we be looking at are we looking at lung MRI, chest CT? Yeah, I mean, there’s always a chest x ray, but I mean, that’s not giving you 100% of the information. I feel like that’s all for within your body, right?
Chris Woods: Lung cancer screening is low dose. Ct So if somebody is in a lung lung cancer screening program, they’re going to get CT scanned. If you see a lesion like a coin lesion or something on a chest x ray, they will send you for a most of the time. Just send you right over for a chest CT. So instead of this projection radiographic image, they’re going to send you to a CT to do tomographic imaging to see that really is a lung nodule and try to assess how big it is and what the extent of it is and precisely where it is. So CT is really the gold standard when it comes to imaging. I don’t think they’re I haven’t read personally about a lot that’s going on in ultrasound in terms of triaging these lung nodules and figuring out what to do with them. I think people are still trying to figure out how to get people in for CT screening because that’s the one that’s been proven to work so far.
Jennifer Callahan: Okay. It’s interesting that MRI isn’t a part of it yet. Just because I feel like Mr. is used, like you said, in breast detection also, too. I mean, I have experience with within my own family. My father had kidney cancer and MRI was the determining factor after the Cat scan that they saw something on his kidney that they then sent him for an MRI or contrast for his kidney that determined that, yes, it was cancerous. It’s heterogeneous instead of homo. So it’s interesting to hear that they don’t that it’s not part of detection for the chest or for the lung.
Chris Woods: And I’m sure people are working on that. And obviously, one of the problems back when I was in decades ago was that we couldn’t we were nowhere near as fast, right? The CT scanner was super fast. And you can see these nodules, especially the wrist, the ones that are dangerous, they do have different hounsfield units. So you can see them pretty clearly in CT. And you do have to fight motion artifacts still when you’re using on the lungs because these patients are breathing. And but I do think that. Both Mr. and Pat. Pat Especially if you have a nodule that’s greater than eight millimeters. Usually one of the one of the options, not usually one of the options in these nodule clinics or tumor boards is, okay, this patient’s got a centimeter nodule. It’s scary. We definitely don’t want to operate on Meredith because she’s got whatever wrong with her. So let’s enter for a Pet scan. And that can oftentimes be the next step in figuring out what to do. But that’s super expensive, and you’re only going to run that on larger nodules because the pet doesn’t give off enough signal. When you’ve got like a 4 or 5 millimeter nodule, you’re not going to be able to really see it in that, right? So that’s where our tech comes in. We’re trying to figure out which nodules look scary and which ones should go to the nodule clinic and maybe bring down the stage at which these cancers are being detected and diagnosed.
Jennifer Callahan: Because guess by the point of something reaching the eight millimeters, I mean, like essentially that’s relatively tiny. But inside the body it’s not right.
Chris Woods: Yeah, and exactly when that switch is going to occur and that nodule is going to all of a sudden start growing and become something that could be metastatic. That’s really difficult to determine when that is. Right now, we just don’t have the tools. So if you detect a nodule and you come back in three months, even if that nodule is malignant. If it if you scan the patient again in three months, only 7% think it’s 7% doing this from memory. Sorry, but think it’s only about 7%. It’s very small number of nodules that will actually show growth like significant enough growth that you can say, Oh yeah, that’s cancer. So just having the patient come back. Isn’t necessarily going to give you the definitive answer whether or not that patient’s got it. You’re going to need eventually you’re going to need tissue. Right. But you don’t want to do that because it’s expensive, hard, and could hurt the patient.
Jennifer Callahan: Right. So then let’s enter in then to to reveal give us a breakdown of the technology that you’ve developed in helping to do the lung cancer detection.
Chris Woods: Yeah, it’s really simple. 30% of the time when you do a chest CT, somebody’s going to have a nodule. Vast majority of those are benign. Right. That if you run the numbers, there’s 15 million chest CT is done in the United States every year. There’s about 5 million. So 14,000 nodules a day are being found right. In the United States. So. Finding them isn’t really the problem we’re trying to solve, right? There’s plenty of nodules being found, like just all over the place constantly. What we’re given. The radiologist is a tool that allows them to click on a nodule. And it gives them a score. And that score has been shown to have clinical value in triaging these nodules beyond what a human being can see. Okay. And beyond just size of the nodule. So essentially what we’ve done is we’ve trained an AI to find what we call radiomic biomarkers, so little markers that are invisible to the human eye that are expressed in the CT scan itself by the nodule that the AI system picks up. And if it picks up a few of them, we run it through a classifier and create a score and. A malignant nodule will tend to have a very high score on what we call the malignancy similarity index.
Chris Woods: And a benign nodule will tend to have a very low score. Okay. None of these tests are 100% definitive, but we can take a nodule in one of our cohorts and one of our clinical studies nodules that are less than 1% likelihood of being malignant. And we can identify those nodules in that class of nodules that are high risk. And those nodules, as identified by our software as high risk, have about a 18% likelihood of malignancy. So that’s a very different conversation to have with a patient, right? We found something really small. There’s less than 1% chance of malignancy. We’re going to have you come back in a year or whatever. Different conversation. When you say, yes, you have a small nodule, but the system flagged it as dangerous. It’s probably a 1 in 5 chance it’s going to have. It’s going to be malignant. You need to come back and you need to come back frequently. And we’re going to spend money to scan you again and again to make sure that this thing is benign and that increase in ability to follow these things and triage them effectively will result in fewer false positives as well as earlier detection. That’s our thesis anyway.
Jennifer Callahan: So I don’t know if this would be too technical to get into, but what are the different markers on the like? Was it the nodules that the eye is picking up to give the MSI score.
Chris Woods: To talk about the specifics? Probably be But suffice it to say we’ve we’ve looked at 14,000 different features of the nodule. Wow. So mathematical transforms and projections, things like that. And then we curated the list down to those that we knew were important. And clinically relevant. And then we built our classifier around that. There’s a patent on the technique. So I’m not saying anything that’s a secret here because it’s the patent is published and issued and the technique looks at specific features. Like I said, there’s hundreds of them that are included in the classifier. But we looked at the ones that are like really the most important in it. They have to do with the shape, especially in 3D of the nodule. We also do features of the surrounding tissue, so the tissue around the nodule. But we don’t include things that might confuse a classifier. Like if you start throwing the whole CT scan in, right, and you treat the eye like a black box, it might say, Oh yeah, people who are of a certain size or weight or sex or whatever are more likely to have malignant nodules because of the training cohort that you feed to it.
Chris Woods: We eliminated all of that black box type of stuff. We don’t let the eye figure out exactly. We don’t let the eye choose any feature it wants to include in this in its determination. We made sure that these were clinically important features. So we curated the list and then we trained it on that. And then we validated on a thousand patient study with the Fred Hutch Cancer Center and Saskatchewan Health Authority in Canada, as well as three other studies. And we’ve proven that maybe the most important thing about the product so far is we’ve proven that it’s generalizable. So you can take this product and plop it into any, we think, any provider out there and it’ll work, it’ll add value and that’s hard for some systems to do. Sometimes they’re trained on one set of data and they work great at that institution where they were trained, and then you try to bring it somewhere else and it doesn’t work. So we’re at the stage now where we’ve proven that you can bring it anywhere and and it’ll work. So which is why we think we’re ready for commercialization.
Jennifer Callahan: Great. So then you’ve had to go through you said it’s patented, but I guess you’ve gone through like FDA approval at this point.
Chris Woods: So we’re in the process of getting FDA. We’ve got our CE mark, so we’re in the market in Europe, just entered the market over there. We have approval in Canada and we have approval in Australia. New Zealand, FDA is next and we’re hoping for early next year.
Jennifer Callahan: Is the FDA criteria so much more stringent than surrounding countries? Other companies I’ve spoken to that require FDA approval. It always seems like America is like the last stop on the approval process for them.
Chris Woods: I think medical devices in general and the European approach, which has been adopted by a lot of other countries, is more like generated data around safety. Make sure that you can recall and then we’re going to let the the clinicians decide whether they want to buy and use it. In the US, we’ve got the safety and effectiveness and both of those are taken, I think, equally seriously by the FDA. So you have a bit of a different bar hurdle to get over. And then of course reimbursement is different everywhere. And in the US it’s especially difficult because of our fragmented reimbursement system.
Jennifer Callahan: Yes. So once you go through your FDA approval process, do you have certain companies that you’re interested in contacting about this or going maybe more towards a health system to get this integrated into their workflow?
Chris Woods: Yeah, we’re actively signing up health systems typically that we’ll we’ll sign up a health system that has the lung cancer screening program in place, maybe has a nurse navigator who’s following up on patients already. Because if you don’t have that, there’s not another reason to improve your triage of lung nodules. You really need to have somebody who’s going to follow up and track these patients. But if you have a screening program in place with everything from community outreach through follow up, we fit right into that. And those are the types of customers we’re trying to sign up. The bigger groups that already have something in place to triage nodules and they want it better. Okay. And we need them to submit for reimbursement because there is a code and there has been a national payment rate set for it. But the way health care works in the United States for Medicare is that you have individual Macs, which are these administrative centers around the United States. You need to get through all of those in order to validate that your lung cancers, you know, that this lung cancer triage software is actually going to get paid. And that’s what we’ll do with our first partners is just prove that out. And then after that, we’ll expand into smaller parts of the community and different types of health systems.
Jennifer Callahan: Have you found a good amount of health systems that have the lung cancer screening in place already?
Chris Woods: It’s evolving. What lung cancer screening did? Was it sort of shined a light? On lung nodules. We do the 15 million chest sets every year. We find all these nodules, but there really hasn’t been a great follow up done at most institutions. But when they when they initiate lung cancer screening, they tend to put in the infrastructure that you need. So then you’re following both incidentally, found nodules as well as nodules found during screening. And that’s been great. I think that that influence of lung cancer screening on health care in general has been really valuable because incidentally, found nodules actually are just as likely to be malignant as those found in a screening program. So putting that infrastructure in place, getting everybody from, like I said, community outreach all the way through the pulmonology department is doing the interventions involved in optimizing. This really will make a dent in health, in the health of a population. I wish it was going faster, but I can’t say the penetration is what we would like to see. I think 6% of all patients who qualify for lung cancer screening now are getting screened. So it’s abysmally low. State by state. It varies. One of the things that we’ve seen is that states like Kentucky have put a lot of money into patient outreach, and it actually does work. Kentucky has way higher. Places like Massachusetts also way higher rates of lung cancer screening than other parts of the country. So if you put effort into it, you can actually really improve lung cancer screening uptake just through patient outreach.
Jennifer Callahan: That’s interesting. I would have never thought that certain states would have had higher rates than others.
Chris Woods: I think a lot of them are. They consider it an important issue in places like where they have mining. You find like Australia, for example, is really serious. They’re rolling out their lung cancer screening program now because they have a lot of miners. And in today’s world, I think we’re realizing, unfortunately, that lung cancer isn’t just a smoking disease, right? Even wildfires can cause people to be at high risk for lung cancer if they’re repeatedly exposed to that. We actually are here in Seattle. Wildfire smoke is not just benign. It’s something that can really cause harm. Firefighters, there’s been some data around them. So we’re identifying these higher risk patient populations that really should be screened as well. And I think that will be one of the things that will evolve eventually. We’ll find genetic markers as well. Sure. And throw people into screening programs based on that. But that’s not the standard of care right now. It’s just people who are smokers get screened.
Jennifer Callahan: Okay. That was one of my questions that I was going to ask you. You’re talking about like the shapes of the nodules and stuff. And I was going to ask if if ethnicity had anything to do with as a marker for detection or that would go into certain diseases. Some ethnicities are higher in it than others, or sometimes, like women are more likely to have heart disease than men. So it wasn’t sure if gender or ethnicity was one of the reasons to get checked.
Chris Woods: I think in the future it will be with a 6% penetration of people who are smokers. The focus has been just get smokers in first before looking at these other subgroups. And there’s two issues with that. One is definitely identifying those high risk patients, but also you need to make sure that your system in our case, we have to make sure our AI works for all of these patients. You need a diversity of patients on many different levels age, sex, gender, race, all those things. When you train the AI and you give it that diversity of patients in your training, it will work on all those patients. If you leave out a subgroup, it may not work as well on that subgroup. So it’s a very important issue for when it comes to the training of an AI system as well as finding those patients that need to be screened.
Jennifer Callahan: No question about the technology and doing the actual scan. So patient comes in for a CT chest scan. They might not necessarily be coming in for lung detection screening, But like you said, incidentally, they’re finding nodules that are there. Is the radiologist having to then open up the application and then apply it to the scan, or is it something that would just already go hand in hand with the scan and your application would pick up? Here’s this nodule and it’s already broken down? Or is it something that the radiologist has to engage to do?
Chris Woods: It’s up to the radiology department, really. If we do automatic detection of the nodule, then we can compute this score automatically and they don’t need to do anything. We just present it to them when they’re looking at the scan. Some radiologists don’t want nodule detectors because every nodule detector is going to generate false positives. So they’re going to have this big list of things that they need to go through and triage themselves. They don’t want to necessarily get rid of all those false positives and have the liability of what if I made a mistake and that nodule comes back. So you’ll find that radiology groups vary in terms of their acceptance of something like a nodule detector. If they have subspecialists, highly trained subspecialists reading all their chest X. Then maybe a natural director isn’t going to add a lot of value to them. But if they’re a radiology group that’s maybe smaller or maybe they have weekend coverage or whatever, that doesn’t. They’ve got more general radiologists and other sort of fill in radiologists reading a lot of chest CT. Maybe a nodule detector would be better for that type of practice. So in that case, with the detector, we integrate right in. If they’re the type of practice that wants to detect them with their eyeballs and then employ our technology after that, we need to integrate with their PACs and that is a job. So it takes some effort on all sides, but it’s not impossible nowadays. It’s a lot easier to integrate with a PACs than it used to be. Okay. So we just need to work with that vendor that PACs vendor and figure out how to integrate in in a way that makes sense. And it’s doable. It’s a solvable problem.
Jennifer Callahan: Now, would they pick and choose, say, possibly the knobs that they see in the scan like this one to to my naked eye, doesn’t appear to be suspicious. So I don’t want to check this one out. But this one over here does. And they just then use the software on, say, like a suspicious looking. Nodule in their opinion.
Chris Woods: Yeah, it’s up to the radiologist, but what you’re saying makes some sense clinically, right? It’s like the patient comes in and there’s six nodules and there’s they’re all clustered and one of them looks bigger and better. And then the rest, then maybe you just want to run it on that. But it depends on the patient. It depends on what they’ve decided to do as standard protocol, The software, from a cost standpoint, it’s the same. Like you can find 100 nodules in a patient and you can use the software 100 times that we’re going to charge you once and you’re going to get paid once. So they can use it as much as they want. But I do think that there does come a point where you check one nodule or two and you’re like, okay, well, this patient’s got something going on here. And what they’ve got going on, these nodules are probably caused by that. So if you’re a radiologist interpreting a case, you can get this good sense of what’s going on with the patient in their entirety that will inform how they use this software. The scariest, weirdest thing is you got this patient looks fine, isolated pulmonary nodule. No idea where it came from. Patient’s not been scammed before. Like, what do you do in that case? And that’s where our product really shines.
Jennifer Callahan: So we were discussing the future of you going through the FDA approval, but beyond that and beyond you getting a health systems and such, where do you see RevealDx going in the next year or two years? Are you looking maybe to branch into other types of cancer detection or is there another avenue that you’re looking to take the company with?
Chris Woods: It’s possible. Yeah. We have looked at other diseases, other types of cancer as well. And there’s also been a lot of movement lately in the industry. People are talking about getting parts of the world. They do double reading in mammo and they’re talking about getting rid of that, only having one radiologist read and I instead of two radiologists reading. And that’s because of improvements in detection and characterization that I have. Nobody really who’s serious in this space thinks that we’re going to get rid of radiologists or even really triage out normals. Yet today, I think for us, we’re a CAD company, which means we’re trying to assist in the diagnosis diagnosis. We’re probably not going to expand into CAD or CAD, which is detection or triage, moving things up and down a work list based on something we’ve found. We’re more of a cat company, which means we’re looking for these hidden biomarkers. We’re using Radiomics. We’re trying to help in the decision of what to recommend for that patient after something’s already detected. So that’s the focus of our company. So we could go into other areas, but we’ll stay focused on CAD X.
Jennifer Callahan: So you said that word Radiomics And I would like you to give everyone a vocabulary lesson like you did for me in case of anyone picked up on on that word their it’s on their website to explain the application. But so I asked Chris to explain it to me further. So Chris, if you wouldn’t mind just giving a brief description of what Radiomics is because it’s definitely enlightening.
Chris Woods: Yeah. The extraction of features that are clinically relevant from using math and science and image processing from different parts of the image, and then feeding them into an AI system to identify what we call biomarkers, radiomic biomarkers that give you insight into what’s going on with that patient that’s beyond what your eyes can see. So so to train a radiomic classifier, for example, what you need is ground truth. And that’s the challenge for somebody. For a company like us that’s doing Cat X is we need to find those exams where we know what happened with those nodules. We know if they ultimately were biopsied and proven to be malignant or they were followed for at least two years and they were shown to be benign because like I mentioned, you can’t just follow them for three months and say, Oh, it was benign because it stayed the same size. So two years or five years follow up. Getting that training data is difficult, but once you get it, you can extract features and of the image and then feed it into an AI system which will then tell you which features are important and which ones are adding value. And we feel like our data and our clinical study shows that our radiomic biomarkers are important in classifying these nodules and they add a layer of value above what size does. So the size of a nodule is important, but we add an additional layer of value that none of those features, if you look at them as a human, they look like noise to an AI system. They’re able to pull out information from there and really give you something that allows the human to perform better than they would without it. So it’s the combination of them looking at the volume with them, looking at the size, the clinical information, the context and the patient putting all that information combined with our score and then being able to make a better and more informed decision about what to do with that patient in that particular case. That’s what we do.
Jennifer Callahan: It’s amazing. It really is. God bless you. Not to be all like, not on you, but, I mean, like, it’s great. I mean, there’s brains like yours out there, too, that have thought this far into something and have put the time and the effort into developing technology like this.
Chris Woods: Wish I wish I could take the credit, but most of the credit for this whole idea really comes from our CTO, Michael Calhoun, who’s our chief technology officer, and really started working on this problem years ago in a serious way. And there’s been several rounds of investment. What I’m trying to do right now is take some of that good work and bring it to the commercial market, because I’ve had the experience of taking some good CAD type work and seeing the impact it can have on patients. It’s pretty profound. I mean, it’s one thing to do something in a lab and then maybe even deploy it in your own clinic, but when you can get a commercial product out there with wide adoption, the impact you can have on patients is orders of magnitude more. So that’s what I’m trying to add to this to this whole equation.
Jennifer Callahan: How long have you been in this process with him then? Because you said he started about eight years ago.
Chris Woods: Yeah, it’s coming up on three years. We had talked. We’re both in Seattle and doing medical image processing. So obviously we knew each other and I had sold a company to Intel. I had started a company and then a worthless company for radiology and sold it. And then I was looking for my next thing to do and the time was right to get involved. So me and actually a group of investors bought Michael’s company from the investors and restarted it, recapitalized it. And it’s been about two and a half years since that maybe coming up on three. And we’re hoping, like I said, to be in the market pretty soon. We’re already in the market in Europe, but the US market with this new reimbursement code is quite an opportunity for us. So we’re kind of heads down focusing on that right now.
Jennifer Callahan: Yeah, it’s great that they’re looking into doing the lung cancer detection, breast cancer. You hear about it all over the place breast cancer, breast cancer, early detection, everything. So it’s good to hear that they’re actually putting time and effort into doing early detection somewhere else.
Chris Woods: I think there’s no denying that lung cancer has kind of gotten a bad rap because, yeah, you were a smoker. You kind of brought it on yourself. I mean, there’s all these dynamics of like whether or not we should pay for screening for these folks and all this stuff. But look, lung cancer hits a lot of people, and I think we’re learning that. And former smokers and the people who are getting screened nowadays, people my age and older, smoking wasn’t like my dad smoked. Right. And he’s a smart guy. He didn’t you know, he was just unaware. You can’t really punish people for their behaviors. Otherwise we’d say any lifestyle behavior is going to cause you to be at risk of something, so we’re not going to treat you. And that’s kind of crazy, right? So I think people are waking up to that and realizing, look, maybe somebody did smoke for a while in their life. Let’s help them. Let’s give them smoking cessation. If they are still smoking, try to get them done. And then second of all, let’s get them strangely. Cancer, a lung cancer screening exam only costs $187. Wow. Right. For Medicare. So it’s not the most expensive thing you can do to try to make a dent in cancer mortality. It’s actually pretty cost effective. Yeah, even compared to other types of screening and ultimately, maybe ten years, 20 years down the road, there’ll be a blood test and you’ll be able to just get screened for all types of cancer. But that’s a big question mark. I think that imaging is for a lot of these screenings. Imaging is going to be ultimately where we end up for quite a while, and that’s probably going to be true for lung cancer.
Jennifer Callahan: Yeah, I agree with you there. I mean, like you said, you could do a blood I mean, maybe you could do a blood test in the future of something like that’s developed. But then you have to figure out, okay, well, see, there’s these types of cells. Where is it coming from? You know.
Chris Woods: Immediately you do it.
Jennifer Callahan: Ultimately, you’re going to be going into some type of imaging because you have to see inside the body. So definitely. That’s right. Chris, thank you so much for taking the time with me today. I really had such an interesting conversation with you and such great developments going on out there for anyone for the future. Hopefully no one in our audience is going to be in that list of the cancer. But if you ever need it, it’s such a great opportunity for your health.
Chris Woods: Yeah, we’ll leave it with this. If you have a lung nodule, it’s found in your lung CT scan. Take it seriously. Right. You may hear that it’s a low risk of malignancy. You may hear lots of things, but definitely take it seriously. Take your own responsibility for following up on that, staying on it, because sometimes they fall through the cracks and you don’t want advanced lung cancer, right? The sooner the better, for sure.
Jennifer Callahan: Being your own advocate in health is number one. So like you said, don’t take anything with a grain of salt. Move forward with it. Chris, thank you so much. I’m hoping to hear about coming out into the US market and getting into the cancer detection. Program. That would be amazing and can’t wait to hear if you’re doing anything in the future. I hope to speak to you soon about it.
Chris Woods: All right. Thanks, Jennifer.
Jennifer Callahan: You’re welcome. Have a great day. You’ve been listening to the skeleton crew brought to you by X-ray technician Schools.com. Join us on the next episode to explore the present and the future of the rad tech career and the field of radiology.
You’ve been listening to the Skeleton Crew brought to you by X-raytechnicianschools.com. Join us on the next episode to explore the present and the future of the rad tech career and the field of radiology.