High Tech: Innovative Perspectives in Craniofacial Care

Bob Underwood (Host): Welcome to Pediatric Frontlines from Shriners Children's, where we explore the best in pediatric care. I'm your host, Dr. Bob Underwood. Today, we're talking to an expert in innovative technologies in craniofacial care, board-certified plastic surgeon from Shriners Children's Chicago, Dr. Chad Purnell. Dr. Purnell, welcome to Pediatric Frontlines.

Chad Purnell, MD: Thanks for having me.

Host: Yeah, absolutely. And this is going to be some really interesting things about technology. So, you recently researched the use of wearable tech for sleep apnea testing. Can you tell me a little bit about how that study came about?

Chad Purnell, MD: Sure. So, one of the cool things about working at Shriners is that we have a huge catchment area. So, I'm taking care of patients from all across the Midwest and even from coast to coast sometimes. And whenever you're trying to be diagnosed with sleep apnea, the key study that you need is a polysomnogram. But that is a little bit difficult for some of our patients to get, because it requires an overnight stay in a sleep lab that can take care of a pediatric patient, and that's not something that's accessible to everyone.

And one of the things that I wanted to do is work with my collaborators to see if we had any way to use something that they could wear at home to help detect sleep apnea. And so, I was introduced to Dr. Hong Yeo at Georgia Tech and he had developed actually already a wearable sensor that had been shown to be accurate with respect to comparison to a polysomnogram. And so, I wanted to put it to work in my cleft and craniofacial population.

Host: And what were the results of this test once you were able to do that?

Chad Purnell, MD: So, the first thing we had to do is just work through some of the logistics of how do we get patients to do these at-home tests. And so, we worked through some of the bugs of having them do tech support over the phone to do this wearable sensor. It's pretty streamlined once you get it working. It's just a sensor that works, it's around the eyes and it gives us a lot of information about the patient's sleep apnea.

And so, what we found was pretty interesting. So, I looked at patients who were going to have jaw surgery. And we know that patients that are going to have jaw surgery, a lot of them end up having sleep apnea. And a lot of our patients did end up having sleep apnea. And what was cool is we were able to look at some of the existing questionnaires that we used to help diagnose people as a screening exam. And what it showed was that patients, they typically take one of two screening exams, the STOP-Bang, or the pediatric sleep questionnaire. And we showed that the pediatric sleep questionnaire was actually much more correlated with the findings on the wearable sleep sensor.

And so, we're going to start using that to screen our patients for sleep apnea prior to any jaw surgeries. And we're looking forward to kind of increasing that collaboration in the future.

Host: Oh, that's awesome. And what are the patient benefits for this type of device?

Chad Purnell, MD: So, a polysomnogram is something where they're in a sleep lab kind of wires from head to toe, and it's a little bit of an uncomfortable place to sleep, right? I mean, it's hard to sleep whenever you have all these different wires hooked up to you, and it's not a very good representation of what a natural night of sleep is. But this is a flexible, wearable sensor that patients are wearing in their own bed, sleeping in their own like normal position. And so, I think it's a lot more comfortable of an exam, and also gives us a better sense of how are they actually sleeping at home as opposed to how are they sleeping when they're in a hospital bed in a lab.

Host: Yeah, absolutely. And do you see potential for the increased use of bioelectronics in your specialty? And I think I know the answer to that. And so, why or why not do you see that as a benefit?

Chad Purnell, MD: Sure. I mean, the key is now we're starting to be able to bring some of this medical testing that would typically be done in a hospital or in a clinic, and we're able to actually bring that to patients at home. And that's good for a lot of reasons. One being just patient convenience. They're getting an exam in their own environment. And also, it's convenient for us, because we are getting better data. We're seeing these patients kind of in their own home environment, and it's really getting us a lot better of a sense of how these patients are doing at home.

So, I think that these wearable electronics are going to really revolutionize the way that we diagnose a lot of things, including sleep apnea.

Host: Yeah. I agree with you a hundred percent. And so, now on to some other really cool tech. So, you presented on acquiring 4D photography in a patient with cleft lip and palate at the 2025 Shriners Children's Burns and Cleft Meeting. Can you briefly explain what is 4D photography and how is it currently being utilized in medicine today?

Chad Purnell, MD: Sure. So, I think to understand what 4D photography is, we'll start at 2D. So, 2D photographs are just your regular photographs that you'd snap with your phone or a camera. We've had 3D photography for a while. So, this is basically using an array of cameras that all take a picture at once from multiple angles. And that actually gives us a full three-dimensional picture of the patient's face that we can rotate and move around, and we can perform measurements on that picture in 3D.

But the 4D revolution kind of is relatively recent. And what it is, is imagine that 3D picture that you're taking with an array of cameras, but now we're actually taking a video. So, we're able to have a full three-dimensional image of a patient while they're animating. So, they're making facial expressions, they're opening and closing their mouth. They're showing us different things. And so, it really gives us not only that extra 3D advantage, it also shows us how are they functioning while they're moving.

Host: Yeah. And that's incredibly important for, you know, the articulation of the joints and everything else that you're trying to capture. So in your proposed application with a cleft lift or palate patient, what exactly do you seek to capture in that imaging, and why is that important?

Chad Purnell, MD: So, one of the things is that sometimes we're fooled by a regular two-dimensional picture. So if you take a static image, you kind of can choose the best static image that represents the way the patient looks in the sort of best light. But there's nothing to hide whenever you're taking a video.

So, what the 3D video really shows us is how is that patient animating, how is that patient moving. And so, we're really getting a sense of how good are our cleft lip and palate results when the patient is moving around. So, this is giving us the ability to determine how different muscle function is happening, how good the symmetry of the face is, when the face is functioning, as opposed to taking just a static picture.

Host: And so, the standard imaging would be a static picture or even 3D, as far as the technique. How is 4D so much better for this?

Chad Purnell, MD: Well, I think that, in cleft lip and palate, the difference is incremental. So, clearly, a 3D image is better than a 2D image, because that's going to give you a sense of how things look with all the different 3D relationships, especially with the relationship with the nose to the lip. But where 4D really becomes important is in some of the other craniofacial diagnoses that I'm treating, things like hemifacial microsomia, where some of these patients have paralysis of some of the muscles on one side of their face. So, a lot of these patients, if you take a single picture of them, you can capture them in a point that shows perfect symmetry. But in reality, as that patients going through their life, they don't have perfect symmetry.

So, I think for patients especially who have facial asymmetries related to muscle function, this is really revolutionary in us kind of describing the problem that they're having and giving us the ability to treat it.

Host: It is just phenomenal, the way we use tech to do that kind of thing. So, at the 2025 Shriners Children's State of the Science focus on burns and craniofacial research. You also discussed an AI-based approach for quantifying changes in an airway due to OGS. Now, what role exactly does AI play in this? And how does it compare to the human mind and what it can do in this kind of instance?

Chad Purnell, MD: So, I think that one of the cool things about AI, is it gives us the ability to kind of give it all of the data, of all of the patients we've operated on and train the AI to help predict what's going to happen with the next patient. And even though I can see in my mind's eye what might happen for that patient, I can't give the patient a picture of what that's going to look like, whereas artificial intelligence actually is going to give us that ability.

And so, the way that we're starting this, we're kind of building this from the ground up once again with my collaborators at Georgia Tech and with Eric Liao, who's one of the Shriners surgeons in Philadelphia. So, with this AI technology, what we're doing is we started by looking at patients that had had orthognathic surgery sent CT scans before and after. And we wanted to start with something that was a little bit more abstract, which was how does the airway change? So, we already know that whenever we're moving the jaws, that the airway changes—the nasal airway and the oral airway.

And so, what the AI first allowed us to do was just to take those CT scans and click on them and basically say it already showed us the airway. So, the AI has already selected the airway for us. The next step is now predicting as we move the jaws what is that airway going to look like? And the next step with this, so as we're sort of training the AI to look at how the airway is going to adapt, the next step is really how is it going to adapt and show us how the face is going to change. Because we already know that orthognatic surgery changes the face, but the interesting thing is that all of these models of how the face might change with orthognatic surgery, they weren't trained on cleft patients. And so, when a patient has a cleft, there's no great model to say, "Oh, if we move the jaws this far, this is how your face might change." We're trying to build that model with AI as kind of the tool that we're using.

Host: A whole new experience of what before and after would look like when you're using AI to help generate images like that.

Chad Purnell, MD: Yeah. I think it's the number one thing my patients ask for, is what will I look like after this surgery? And right now, I can see it in my own mind, but I can't show them very well.

Host: Right. It is amazing to me. So, do you have any reservations about AI advancement in medicine? And if so, why? And if not, why not?

Chad Purnell, MD: I mean, I do. I think AI is really exciting, but like any exciting technology, there's always the potential for this to be used in the wrong way or to be overused. I think the most important question right now is just as we're utilizing AI for patient data, we have to make sure that we're protecting that patient data. So, that's one of the most important things that I've worked with with Georgia Tech, is that we are ensuring that even though we're feeding actual patient data into the AI, on the other end, you're not going to be able to pull out any of the original data that was placed into it. So, we want to make sure that all of our patients who their data is being used to train the AI, we can't just pull that back out and show that patient to another person.

So, that's really the first concern with AI is the patient data and patient privacy. So, we're making sure that we protect that really well. But I think, you know, as AI becomes more powerful, there's always concerns about is AI going to take over some of the decision-making for physicians. And I think realistically, especially as a surgeon, we're a ways away from that.

I think that AI right now, it's a helpful tool and it certainly saves time. It can give us a better sense of some of the options for patients. It can give us a better sense of diagnostic potential, but we're not quite there as far as the AI actually replacing a physician. Especially as a surgeon, at the end of the day, I have to go into the operating room and operate on these patients. And so, the buck does not stop with AI if that makes sense.

Host: Yeah, I agree with you completely as a physician myself. So, what other high tech advances are you excited about when it comes to craniofacial care?

Chad Purnell, MD: So, this is a very exciting time for craniofacial surgery. No question. We've kind of seen multiple advances in craniofacial surgery. The first big advancement was virtual surgical planning, where we basically could plan out our surgery in an engineering environment. And that helps us kind of make guides and helps us to make things sort of more accurate in the operating room.

The next step was kind of patient-specific implants. So now, I'm routinely 3D printing custom implants for patients, custom plates for patients, and you get sort of that one of one kind of custom patient solution.

The big jump that we're about to make in craniofacial surgery is virtual reality and augmented reality. We are very close, not quite there, to the point where in the OR, I'm going to be able to put on augmented reality, extended reality glasses, and I'm going to be able to see an overlay directly on that patient of their bone structure. And it's going to help guide me during surgery.

Now, the problem with craniofacial surgery is you need a millimeter of accuracy. So, you can only be off about a millimeter. And the current augmented reality technologies that are available are still centimeter level accuracy. So, it's not quite there. But the thing is that there's an exponential increase in this technology. And so, I would imagine in the next two, three years, we're going to be seeing that millimeter level accuracy that's going to be required for craniofacial surgery.

Host: Wow. And what are you working on now, and what's next for you?

Chad Purnell, MD: So, you know, for me, the thing that I'm most excited about, that I'm working on is something a little bit off the research path and more on the clinical path. So, some of the most difficult patients that we treat are patients that have Treacher Collins syndrome, who also have a small jaw and a small airway, but also don't have normal TMJ or temporomandibular joint. And these patients have been notoriously very difficult to treat.

But I have—kind of standing on the shoulders of some of my mentors—developed a treatment plan that I think is working pretty well for these patients that I call distract and fill, where I'm actually pulling the jaw forward using a halo distraction device and then replacing the TMJs with a custom temporomandibular prosthesis.

And what I'm really excited about is these are patients that before they'd have to choose between getting the right bite, getting the right airway, and getting the right facial appearance, and now we're making fewer and fewer compromises with these patients where we're actually getting them the jaw function and appearance that they need, as well as getting them removal of their tracheostomy and improved airway. So, I am really excited about treating more patients with this protocol.

Host: Yeah. And that's phenomenal. And you've talked to a number of things of why Shriners Children's really brings access to care and types of care that many patients, many children would just not otherwise have access to. So, thank you so much for sharing your expertise and experience with us today.

Chad Purnell, MD: Of course, thanks for having me. I appreciate it.

Host: Yeah, it's been phenomenal. Thank you so much.

Chad Purnell, MD: Thanks.

Host: And for more information, including the full range of care disciplines, please visit shrinerschildrens.org. And to hear more Pediatric Frontlines episodes, please subscribe wherever you listen to podcasts.