Derm-it Trotter! Don't Swear About Skincare.
Feeling frustrated or overwhelmed with everything skin? Does the skinformation overload make you want to swear about skincare? Join Dr. Shannon C. Trotter, board certified dermatologist, as she talks with fellow dermatologists and colleagues in skincare to help separate fact from fiction and simplify the world of skin. After listening, you won’t swear about skincare anymore!
Derm-it Trotter! Don't Swear About Skincare.
Beyond the Biopsy: How AI Is Changing Skin Cancer Detection
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
A biopsy is still the gold standard for diagnosing skin cancer, but it is far from perfect. From the AOCD meeting in Orlando, Dr. Laura Ferris, Chair of Dermatology at the University of North Carolina, joins us to break down how skin cancer detection is evolving and why tools like dermoscopy, mole mapping, and total body photography are changing the way dermatologists evaluate suspicious lesions.
We also explore the growing role of AI and FDA-cleared technologies like Nevisense and Dermasensor, plus the challenges of access to dermatology care, especially in rural communities. From smarter screening to reducing unnecessary biopsies, this episode looks at what the future of skin cancer detection could really look like.
Why Skin Cancer Diagnosis Is Evolving
SPEAKER_02And we want to make sure that we're not, you know, doing biopsies that are unnecessary as well. Light normally hits the skin and scatters back, but it helps us to sort of let the light penetrate deeper into the skin. We could see structures that we just can't see with the naked eye, you know, analyzes in real time those images and says, is this likely to be skin cancer? Yes or no. We can't send everybody to a dermatologist to do a skin check. And think about it.
SPEAKER_00Like when you are looking for breast cancer, welcome to Dermit Trotter, Don't Swear About Skin Care, where host Dr. Shannon C. Trotter, a board-certified dermatologist, sits down with fellow dermatologists and skincare experts to separate fact from fiction and simplify skincare. Let's get started.
SPEAKER_01Welcome to the Dermitrotter Don't Swear About Skin Care podcast. We're coming at you live from the American Osteopathic College of Dermatology AOCD meeting in Orlando, Florida. And we're going to talk about something that a lot of Floridians are familiar with, skin cancer, but a lot of you listeners out there too. We've got a special guest here on today's episode, Dr. Laura Farris. She's a board certified dermatologist and chair of dermatology at the University of North Carolina. She specializes definitely in skin cancer and pigmented lesions, which most of you know is simply moles. So I'd love to have her on the podcast to talk more about skin cancer and how the diagnosis of skin cancer is changing. So welcome, Laura. Thanks. Great to be here. So, you know, I think people think of skin cancer the traditional diagnosis when we're looking at it that, you know, we do the biopsy, we evaluate it. I'm gonna have you walk me through kind of how that occurs, but really where we're evolving and kind of complementing that or potentially even moving away from that method.
SPEAKER_02Yeah, sure. So, you know, I think that traditionally you think of it as you go to see your dermatologist, they look at all your skin, they pick out the moles that look funny, they do a biopsy, it goes to pathology, and you get your answer, skin cancer or not. Um, and that's still kind of the at the heart of what we do. I think that where things are moving, or as we've looked at this a little bit more critically, to say, um, you know, what are the downsides of that? You know, one downside is you might miss a subtle skin cancer. And then the other downside is you can end up biopsying a lot of benign stuff in order to really try to avoid missing skin cancer. So that's where I think technology can be really helpful. A lot of us use a dermatoscope, sort of a handheld microscope that illuminates and um, you know, we are we're trained to look at special features that can help us be a little bit more discerning about what we choose to look at further or do a biopsy. And then there's also technologies that we can use, anything from things that look at a single lesion if we're not sure about it, and give us more data, or even things like whole body imaging, where we can have photos of the body of somebody's whole body, mole mapping, we call it sometimes, where your dermatologists can then, you know, see what a mole looked like, see what you looked like six months ago, one, two years ago, what you look like today, and see what's new or changing.
SPEAKER_01So, what do you think is driving? I mean, obviously, technology we know will change how we practice medicine, but do you think this drive to find better ways of diagnosing skin cancer, do you think it's because patients are reluctant to maybe have biopsies? And some of you might be saying, well, are we not good at diagnosing skin cancer, you know, as dermatologists? What do you think is driving this whole need for this?
SPEAKER_02You know, I think there's a few things. So um, one, I do think patients want all, you know, this is the era of personalized medicine. Patients want to know, you know, more, not just this is your opinion, you're going to do a procedure. I think, you know, particularly in cosmetically sensitive places, um, people don't want the scar of a biopsy. On the other hand, they really don't want to miss their skin cancer. So I think that there is great interest in um having a little more data before we make these uh decisions. You know, I also think that as we're very conscious about um costs in medicine. And we want to make sure that we're not, you know, doing biopsies that are unnecessary as well.
SPEAKER_01Yeah, I think that's a big piece for me because I mean, you know, there's a certain, we always talk about how many benigns we should find to something that's actually a skin cancer. But at the same point, if I could save some of those benigns and potentially not biopsy a patient, and if I could maybe get some help identifying that spot, I mean, I do have confidence, I think like most dermatologists in our clinical skills, and I love using dermoscopy, but I do think there's the ability to have technology that could help us. Now, you talked a little bit about uh whole body or full body photography or total body photography.
Mole Mapping And Change Over Time
SPEAKER_01What does that really entail for a patient? What would they have done?
SPEAKER_02So, you know, this is something we've been doing for decades now. And so at the very most simple form, it is you go into a room, there's a medical photographer, you are undressed to either completely undressed or just down to undergarments, depending on the institution or the office and the comfort level. And, you know, at the very basic kind of low-tech level, um, a good photographer takes very high resolution um photos in maybe 16 to 24 different positions. Um, and those get uploaded into, you know, they could be printed out and put in a photo book or uploaded onto a server. And then your dermatologist or um, you know, provider will end up looking at those photos to say, if they have a lesion that's suspicious, is this mole new? Is it changing? Or did it look exactly like that years ago? Now with technology, you can imagine. We can start, you know, some of these systems are a little bit more sophisticated. Um, you can, for example, use AI or computer vision to try to have the computer find those lesions. And I think that that is really still in its infancy, but a really exciting direction that this field is going to be taking. Um, you can also do things like match it up with the met we know we talked about dermoscopy briefly. So, you know, you could click on your dermatologist, could click on a mole, you could blow it up, see what it looked like clinically at a, you know, larger, better resolution, and then also see what the dermoscopy looked like because just like moles change in size and shape, they also change their dermatoscopic features change too. And change is really an indicator that something's going on. A lesion that is not changing in size, shape, dermatoscopic features over years. It may look bad, but we're pretty confident that that's not going to be a melanoma. The lesion that is showing kind of significant change, even if it doesn't look too bad to the naked eye, that's one that we want to take a closer look at.
SPEAKER_01Yeah, so it's, I mean, it's key that stability factor, right? Like people always think we we're ready for change, but for moles, not the case. Now, you touched on dermoscopy, and we've mentioned that a few times. I think patients may not know what it is. Depending upon where they've gone, they may have had it done, not realized they were having dermoscopy on exam. And then some of our colleagues may not even utilize it within an exam. Do you mind kind of talking about what dermoscopy really is and how it can be helpful?
SPEAKER_02Sure. So you may have, you know, seen your dermatologist and they have this, you know, little handheld tool and they look, they put it up against one of your moles and they look through it, and that's dermoscopy. So what it's doing, there's kind of contact and non-contact, but many of us use contact, which is where um they may wipe your skin with an alcohol swab, they put it this um what looks like a handheld microscope onto the skin. And when we look through that, um, we one, it does magnify everything so you get a better view. But two, and you know, light normally hits the skin and scatters back, but it helps us to sort of let the light penetrate deeper into the skin. We could see structures that we just can't see with the naked eye. And to your point, training is so critical. If I put that in the hands of, you know, a primary care physician who's never done it or a patient, they're not gonna know this is a feature that really tells me that's benign, or this one makes me think it's a skin cancer. So it is the tool, but it's really the training that goes along with it as well. And there are studies that show in the appropriate patient, the appropriate with the appropriate training, this can help us find more skin cancers, but also biopsy fewer benign lesions.
SPEAKER_01Yeah, I like to think of it as almost like x-ray vision for the skin, the best we have maybe to a degree when you're utilizing, you know, the clinician's interpretation with a tool so we can see below the surface and maybe get an idea of what's happening for patients. But I love that
AI To Expand Access To Expertise
SPEAKER_01too. So, I mean, how do you look at, you know, the role of you know, artificial intelligence and then diagnosing skin cancer? Because I think this is where we've talked about some of the basics, you know, total body photography, dermoscopy. I think a lot of us as dermatologists are kind of familiar, may use that, you know, total body photography may not always be accessible, you know, to patients and cost being a factor because it often is not covered by insurance. But do you feel like the role of AI is, you know, and I think people are wondering this and even out there, is that going to be better than my dermatologist? Or do you feel like, no, this is something to enhance its more augmented intelligence? Or how do you see this really fitting in a role in how we practice and help patients?
SPEAKER_02You know, I think one of the things that you said that really like resonates with me is access, right? So, you know, your listeners may not know that you are like a nationally recognized um, you know, expert in skin cancer diagnosis. You've worked at huge cancer centers, right? But you are kind of a rare commodity, right? So everybody doesn't have access to a dermatologist who has devoted so much of their career just to learning things like dermoscopy to understanding pigmented lesions. So I see AI as a way to sort of bring expertise to areas that um where you know it doesn't exist. If you look at the data, if we just said everybody could just see a dermatologist once a year, they would be able to find the skin cancers, we'd be fine. I mean, A, if you took every dermatologist in the country, um, they would all have to do like 80 skin checks every day, five days a week, 52 weeks a year to even think about providing that access. That is not feasible. We got to take care of lots of things. Um, and B, everybody isn't gonna come at this with the same level of expertise. So where I think AI can come in is that it can help to sort of bring a higher level of expertise across the board. So um, you know, so that is there are many studies that have shown that there are tools where they have dermatologists-like um, you know, sensitivity or ability to find skin cancer. So one, I think it puts that expertise in the hands of lots of different providers, maybe primary care providers, um, our um physician, uh, physician assistants and nurse practitioners. Um, so one, I think it does that. Two, geographically, everybody doesn't have access, right? So we all practice, you know, I practice in a pretty um, you know, a non-rural area, but I live in the second, the state with the second largest rural population. We don't have dermatologists in the areas where lots of our patients live. So I think technology is something that can help us bring dermatologists like sensitivity and specificity to maybe primary care providers. So I think that it's really gonna be at the level of access and sort of bringing everybody to a higher level. Not that this is something that, you know, you, Dr. Trotter, is gonna make you a whole lot better, but I hope it's gonna bring Dr. Trotter level expertise to a lot more physicians. And I love that.
SPEAKER_01Because I do think that's a fear, you know, amongst probably a lot of dermatologists listening and think, oh gosh, like are we gonna have a role anymore? But you're right, that access point is important. And I think too, it can help patients, you know, understand that we're evolving with technology we're utilizing in clinics to help identify skin cancer earlier.
FDA Cleared Tools In The Clinic
SPEAKER_01And now I think what's fascinating is we already have two FDA-approved devices here in the United States, which some people may not even realize or even be implementing into their clinic, or patients might be wondering, gosh, well, should I be going somewhere where they actually utilize those devices? So it's gonna have you tackle kind of the two couple of these. Do you mind talking a little bit about Nevisense and Dermosenser? Sure. So I'll start with Nevisense.
SPEAKER_02So this is an FDA-approved device that is approved for use by dermatologists to help to make a biopsy decision. And so this does identify both melanoma skin cancers and you know what we call keratinocyte carcinomas, which are generally basal cell carcinomas, squamous cell carcinomas. Um, you know, you might say, why does a dermatologist need this? Well, interestingly, everything we've talked about up till now is, you know, using our eyes, either alone or with things like a dermatoscope. This looks at a totally different, you know, feature that we can't see, which is essentially the conduction of an electrical current across a lesion. And it's not something the patient feels, it's a teeny probe. And so, you know, I can't see resistance, you can't see resistance, but you know, um, malignant cells conduct electricity basically different than benign cells do. So, one, I think it's interesting because it lets us see a feature that we don't see with our eyes. And so the idea here is that this would help us to make a better biopsy decision, not to miss things like melanoma, not to miss even things like basal cell carcinomas that are more common, even if they're less deadly, and to find them at a smaller phase when maybe the surgery will be, you know, smaller with a smaller scar. Um, it also has, you know, a pretty decent what we call specificity, so that it doesn't just say biopsy, everything. So I think that this can give a dermatologist a tool that they just don't have, even with the visually based ones that we have. Second one being dermasensor. That's kind of interesting. This is FDA cleared. It's a newer tool. This is actually not intended for use by dermatologists, it's intended for use by primary care providers. And it is to help them decide who needs to see a dermatologist. So we know our your uh listeners probably um often have found, gosh, I think I need to see a dermatologist, and they find the wait's three months, four months, five months. So, really, this is a triage tool. This is sort of getting at the access. Um, can my primary care doctor put this tool on the lesion I'm worried about or they're worried about? And if it says, you know, low risk, it's benign, maybe you don't need to see a dermatologist. But if it says high risk, maybe that's something that can actually help us to sort of figure out who needs to get faster access. This again doesn't use visual features. It uses, you know, something that we call elastic scattering. And so it is um looking at, you know, the a light that is going in and saying, you know, this scatter, the blood vessels scatter like this, the the um connective tissue, you know, the melanocytes scatter, and then they look and you know determine does this look more like a skin cancer or more like a benign lesion? And again, it's evaluating for all types of skin cancer, not just melanoma.
SPEAKER_01And have you noticed our, you know, does there seem to be an increasing interest, you know, from primary care dermatologists to bring these devices to their offices? I think that there is.
SPEAKER_02And so um, I think, you know, one, patients want to see that we are, you know, up to date using the latest technology, right? We're using the latest drugs, we have the latest, you know, in our big centers like PET scanners or CT machine um scanners. So, you know, we also want to have the latest technology in terms of taking care of our patients and their skin health as well. And, you know, in the primary care setting, um, they are just hungry for more help with taking care of um patient skin concerns. We know that that's maybe 40% of what patients will come to their primary care doctor for. They are they don't receive the level three years of training like we do. Um, they may have a one-month rotation if they're lucky. So these are ways to help them, you know, provide their patients more direct care instead of just referring, referring.
SPEAKER_01And I think they're more practical. You know, we talked about germoscopy, it's a fantastic tool, but even as a clinician who uses it all the time, it takes a lot of training and confidence and just experience to utilize it. And, you know, that's not always realistic, especially for a primary care physician when they have so many other things to be managing. To have devices like this, like you said, it improves access. And really for those that need to get to Durham, they can. And it allows them, I think it empowers them to provide dermatologic care in a way that they never were able to do before. Now, here in the States, obviously, we have a couple devices. What about elsewhere? Because obviously this has been an interest internationally in other countries to create new tech and devices. Are there any that you feel like that have shown some merit that maybe should be considered here or you know, some that are on the horizon as well?
SPEAKER_02Yeah, so I think that there um are a one that I think you know really stands out to me is there's a company in the UK called Skin Analytics, and they have this device that they call Derm. And we actually did some of the preliminary clinical trials for them that we did in the US. So and this is interesting in that it's so accessible because it's really a computer program, but it sits on a smartphone and um you connect a basically a small dermatoscope to the smartphone. Um, you take a clinical image, meaning um one without the dermatoscope, and then you can also take one with the dermatoscope, and it, you know, analyzes in real time those images and says, is this likely to be skin cancer? Yes or no. And it kind of goes through a stepwise process. It says, all right, number one, we don't want to miss a melanoma. Is it a melanoma? Yes or no. No, all right, is it a squamous cell carcinoma? Is it a basal cell? And it then says, it then sort of acts as a triage and says, this lesion needs to go see a dermatologist or this one does not. Now, when they studied it, they actually had a dermatologist in the background review every image. Now they are using this autonomously and they are saying, um, all right, this has a low risk score. That patient does not need to see dermatology. And um I, you know, in the studies that they presented, about 25 to 40 percent of lesions were considered low risk by the device. When a dermatologist looked, none of them ended up being skin cancer. So they are saying, don't send that patient. So that is truly going to improve access, um, hopefully, and they are gonna they're gonna study this in real time. So that what they found was they had even just in the UK, way smaller than the US, over a million referrals for um to dermatology for suspicious lesions. I think less than 6% of those actually ended up being skin cancer. So it's like, you know, can we do something to really make these more high yield? You know, stay tuned to see what happens. But I think it's gonna be really interesting.
SPEAKER_01Do you do you think a device like that would make its way here as well over time? Yeah.
SPEAKER_02I think that it could, yes. And so I think that, you know, potentially dermosensor is trying to fill that role here. Okay. Um, but you know, in happening at the primary care level. But, you know, hopefully there's room for more technology to do this too.
SPEAKER_01When you've looked at these devices, and I know they like to look at kind of maybe comparing accuracy of, you know, a dermatologist maybe versus a primary care doc versus the devices, how are these devices performing, you know, compared to us diagnosing skin cancer?
SPEAKER_02So I would say um if you look at most of the devices, and that can include, you know, things that are not on the market, but AI tools. So in general, what you'll find is that they have a higher an equivalent to higher sensitivity as a dermatologist, meaning that um they will find, you know, the skin cancers we find and maybe some additional ones, where it sometimes you know falls off as in the specificity. So we may be more comfortable saying that's benign, I don't need to biopsy. So a lot of devices will say, do more biopsy, but they miss more. Now, those are all kind of tunable features. So, you know, you can set, you can set your threshold for biopsy um, you know, lower or higher. So you may say, I'm gonna make it really high, it's gonna find every single, even very early skin cancer, but you're gonna end up doing more biopsies. You can also say, I really want to, you know, not be biopsying as many benign things. I certainly don't want to miss the more advanced cancers, but you know, something like an early basal cell that's gonna, you know, show its reveal itself and not gonna cause any harm if we delay biopsy a year, there's probably some, you know, tweaking in there to make sure that we are getting the bad ones, but not biopsying everything as well.
SPEAKER_01So if there are dermatologists intimidated by the technology or maybe that don't trust it, what what would you say to them to reassure them if they're thinking about bringing it on board?
SPEAKER_02You know, I would say like don't change a lot of what you're doing in that if you know that's a skin cancer, you don't need a tool. We're gonna do the biopsy. Um, and then, you know, think about the patient in front of you and their risk level. I think that's another thing is, you know, think about the risk level of the patient. Think about the risk assessment that you have for that one lesion. If you're 99% sure it's benign, you probably don't want to, you know, use a tool that may make you, you know, say, well, it's it's two times more likely than I thought that it's going to be. I thought there was a 0.5% chance there's it's uh I got a positive result. I should double that. It's a 1% chance. If that's really what you're doing, you're still probably not going to biopsy something if you think there's a 1% chance that it's a basal cell. So just think about that. Think of, you know, think carefully about how you're using these tools. Um, and then think, you know, what is the great, what am I trying to achieve? Is this something where I really would like to avoid a biopsy? If this can reassure me, I think the rule out feature that I'm on the fence, maybe it's a skin cancer. Most of these are much better at saying if they say it's not, um, you can probably have a much greater confidence that it's not and avoid the biopsy when you really, the patient doesn't want it, you don't. Want to do with the areas concerning it's a high risk of infection, things like that.
SPEAKER_01Well, and and then for going on the patient level, if they're intimidated, but they're like, I'm not going into the office. What
Phone Apps Limits And Future Screening
SPEAKER_01about now what we can get on our phones, right? There's an app for that. So I wanted to get your thoughts because there's a lot out there. I was just trying to look at even from when I last looked to see, but oh, patients come in and they're utilizing these apps, and these apps are telling them one thing versus the other. Are there any out there that you feel like that are actually validated that you would recommend versus do you feel like this is a lot of uh, you know, just unsubstantiated claims that these apps are making?
SPEAKER_02I I really feel like the majority of them are unsubstantiated claims, right? So, you know, when we talk about these devices that have been FDA approved, you know, they had to go through a process of doing a prospective clinical trial where they, you know, had they enrolled lesions and patients and then had everything biopsied and then followed up and they can give you numbers. These apps don't do that, right? I mean, you wouldn't, you wouldn't take a drug that didn't go through a clinical trial where, you know, experts in the field have looked at the data and said, you know, this is, this works and this doesn't. And here's the benefit, here's the risk. Let me quantify the benefit, let me quantify the risk. You know, apps that that are out there do not go through this process. And so I really don't feel like there's anything I'm comfortable telling my patients, just use this.
SPEAKER_01Okay, because I think that's probably the scariest thing that I see now that people want to do this at home, monitor lesions, not come in, or with telehealth now. You know, I always get some blurry image and do I need to come in? And it's always yes. I mean, my staff always knows that will always be the response because I just can't simply tell. And I do try to tell them these are more for fun, but I don't take them seriously, and I don't think they're worth the time right now. And we have better devices, you know, to actually look at that. Or obviously, you know, your dermatologist to evaluate that where you can trust it. So, with all the evolution of what we're seeing technology-wise, like where do you think we're headed into the future? How, how do you predict, you know, what devices are to come, how we utilize things? Like, what's exciting and on the horizon?
SPEAKER_02Yeah, you know, what I think is going to be really exciting is the advances in the whole body imaging. So, um, you know, like I said, we can't send everybody to a dermatologist to do a skin check. And think about it. Like when you are looking for breast cancer, you go to a radio, you go to a site, they a techni, a technician does the imaging, then a physician on the back end looks at it, and really AI does a lot of looking at the images now. And then you go see a physician when there's something of concern. So there's a little bit of a separate, you don't have a physician, you know, palpating the breast on every patient. That's not time efficient. Um, it's not the right way to do it. And I think we got to start thinking about skin cancer screening the same way. So I love the model that um, you know, somebody who's a medical assistant or a nurse or a pharmacist could take these images. Um, they could take them, you know, every six months, every year. Um, AI and technology could then find what's new or changing. If it's not changing, we don't really care. If it's new or changing, then a physician can look at that image. Um, AI would have told that nurse, get a dermatoscopic image of that one because that one's different. Then you could look on the back end, use your expertise. And then only when at that point we say, you know what, that's concerning enough. Let's send that patient into the dermatologist. There are a few companies that are working on tools like this. I think that's going to be really exciting because I can put that in rural North Carolina, you can put it in rural Ohio. Then, you know, one out of maybe 25 patients has to make the trip in to see us. Or we have mobile units that can go to each area once a month, and then we only see the patients who need us the most. To me, that's where it's going, and I think that's gonna be really exciting.
SPEAKER_01Well, I think, you know, AI is something we're gonna all have to embrace, right? Whether you're a patient or a physician or other healthcare provider, it's here to stay. It's just how we utilize it and make the practice of medicine better that counts. But I want to thank you so much, Laura, for coming on the podcast. It was great to have you here and share your expertise with us today. Thanks for having me on, and thanks for doing this great service for our whole community. Of course, happy to do so. And stay tuned for the next episode of Dermotter Don't Swear About Skin Care.
SPEAKER_00Thanks for listening to Dermot Trotter. For more about skincare, visit dermittrotter.com. Don't forget to subscribe, leave a review, and share this podcast with anyone who needs a little skincare sanity. Until next time, stay skin smart.
