Dr. Bob Underwood (Host): This is Pediatric Insights: Advances and Innovations with Children's Health, where we explore the latest in pediatric care and research. I'm your host, Dr. Bob Underwood. Today, we're exploring how advanced imaging and 3D modeling are changing the way surgeons plan and perform complex craniofacial procedures in children through the Analytical Imaging and Modeling, or AIM Center at Children's Health.

Joining us is a plastic and craniofacial surgeon and Founder of the AIM Center at Children's health and professor at UT Southwestern, Dr. Alex Kane, and an imaging scientist and Research Director of the AIM Center at Children's Health and Research Associate Professor at UT Southwestern, Dr. Rami Hallac.

Dr. Kane, Dr. Hallac, thank you for joining us today.

Dr. Alex Kane: Thank you.

Dr. Rami Hallac: Thank you so much.

Host: So before we really get into the partnership discussion, can you give us a little background on the AIM Center? What exactly is the AIM Center?

Dr. Alex Kane: Well, it's an acronym. It's an acronym that stands for the Analytical Imaging and Modeling Center. It is something that is not brand new. I was recruited here from my previous location to build an academic plastic and craniofacial program. And as part of that, my vision was to have something like the AIM Center. My academic background has always been in imaging and in imaging research pertaining to cleft craniofacial and pediatric plastic conditions.

What the AIM Center aims to do, pardon the pun, is to bring in basically three components. We want the AIM Center to assist with patient care, and that can be broadly defined. We want it to help with research. It's generally going to be outcomes research of some sort, understanding anatomy or doing functional outcomes and also to assist interested parties in intellectual property kind of development.

It's important for the audience to understand that the program was not created as a unique resource for plastic and craniofacial surgery. The idea would be or has always been, and is, that it's there as a core facility for all pediatric researchers and clinicians to do all of the things that I've just enumerated. Meaning it's not just for our specialty. It's a generalized place that one can conduct these types of activities.

Host: I think that that's absolutely fascinating in really applying the technologies. What kind of new opportunities does this create for studying and treating for your specialty, complex craniofacial conditions?

Dr. Alex Kane: Well, I think it has come in all of those bins. I think patient care has been a very important part of the way in which we utilize the AIM Center. There are common conditions that we see, and sometimes there are extremely rare conditions that we see.

For the ordinary conditions, often we don't need to utilize the abilities of the AIM Center. But let's say that we have a patient with a very complex problem, and I could give examples. A particular type of complex facial cleft, not an ordinary type of facial cleft. a cleft that would involve both mandible, maxilla, zygoma, going up into the orbit. Yes, you have the ability with modern CT scans to look at images on a screen. But if you wanted to define particular anatomy that may be important for the surgeries that you need to try to assist these patients, then it can be very helpful to have a physical model in which you can look at these problems and hold it in your hand and either do mock surgery on or sort of think about the surgery with.

You can isolate particular areas of concern. You can look at, for example, the globe in the orbit in a much more tactile and intuitively obvious way. That would be one example, and that's only with the 3D printing aspect of things.

Host: Yeah. And I was going to get into that because the AIM Center uses 3D imaging, 4D motion analysis, artificial intelligence, computational modeling. How are these tools being used to integrate into plastic and craniofacial surgery? And what can they show you that conventional approaches may not be able to?

Dr. Rami Hallac: Yeah. So, I think the simplest way to explain it is that traditional measurements give us a few important numbers, but they don't always tell us the whole story. With 3D imaging, we can look at the full shape of the face and head. We can see subtle differences and asymmetries that might be hard to capture with just few measurements.

Then, with 4D imaging, we add movement. So, we're not just looking at a still image. We can see what happens when a child smiles or speaks or even move their face. With AI, we add another layer. So for example, we can look at large amount of images and we identify patterns that would be very difficult to use or measure manually. It can also learn directly from the images and recognize features and patterns. So overall, these tools give us a more complete picture of anatomy and function.

Dr. Alex Kane: And how that integrates with the clinical kind of situation, let's take something that we commonly treat a very common diagnosis that encompasses about 50% of our clinical work would be in trying to help children and their families with cleft lip and palate. The traditional outcome measurements for a child with a cleft would be somebody trying to judge your outcome. Let's say you're trying to see how well you're improving their appearance, and you're trying to judge one repair versus another in terms of appearance.

Well, if you look at the outcome with a single static image, whether that be a photograph or even a 3D image, you may be leaving out some very important signals about outcome. A smile is a very important outcome from a cleft lip repair. And unless you can sort of understand what a normal smile looks like in terms of the dynamism in a very quantitative way, you're missing out on a potentially very important thing to a living individual who has had this condition. And this is why the dynamics can be tapped and quantified with these higher level morphometric kind of analyses.

Host: It's just totally fascinating and just wonderful application of the technology. So, how are you using these technologies across different craniofacial conditions? Can you give us an example of a patient or a condition where this collaboration has really made a meaningful difference?

Dr. Alex Kane: I think it's important to say that not only is there application in plastic surgery, but we've utilized these same kinds of dynamic studies, 4D movement functional outcomes across different specialties. It could be gait analysis for the physical medicine and rehab group. It could be looking at how patients with autism move and do basic movements. So, this laboratory, if we call the AIM Center an a laboratory is a place where you can assay and sort of get into these sort of outcomes in a way that you can't using conventional outcome measures.

And the other important point to understand is that none of these imaging modalities that are employed in the lab are X-ray based, meaning we don't apply any sort of X-ray imaging. What we do here is we apply only light-based, conventional light-based imaging techniques, so there's really no cost or danger to the individuals undergoing it. This allows us to take more frequent measurements. It would be great if there was no issue with taking CTs serially on children. We know that that's not the case. We know that there are harmful potential after effects from getting CTs frequently. We don't want to increase the cancer risk from ionizing radiation. With these kinds of technologies, what we can do is, every time they come in for a visit, and because the lab is right next to the clinic, they can go in and get whatever sort of imaging they want. You can take a 3D image of their face every single time they come. This allows us to have a richer data set to allow us to look at growth and our outcomes, which ultimately are related to how the children grow. And so, you have way more data-rich timeline to which you can then apply these other strategies for looking at them.

Dr. Hallac is very facile at developing very bespoke methods for looking at how things grow and change over time in three dimensions. Yet, with AI, the LLM is opening up doors to allow us to do things that we haven't even thought possible before. And it's coming very quickly. We delved into AI, our toe was in that a long time ago already. But that was with writing our own algorithms. Now, you have machines writing their own algorithms, and who knows what the future's going to hold in that respect.

Host: Yeah. That's fascinating to even try to imagine. And for children with complex craniofacial anatomy, surgical planning can be highly individualized. So how does this advanced imaging help you better understand the individual child's anatomy and plan their treatment?

Dr. Alex Kane: Well, there are certain anatomic structures as you know, that are much more at risk than others in any particular surgery anywhere in the body. So meaning if you're a heart surgeon, the vessels supplying the heart or the nerves supplying the heart are very important.

In a craniofacial situation, you're very concerned about harming adjacent structures, brain, vision, function of the oral organs, these types of things, airway. And so when you have a 3D image or even a 4D image showing us what's happening with these anatomic areas, you can avoid harm by understanding what's at risk in a much more tactile and intuitive manner.

Host: Yeah, completely. Now, you've already talked about how the partnership with the AIM Center is allowing you to study anatomy, function, and outcomes in ways that you couldn't before. How might that research ultimately change the way you're treating patients?

Dr. Alex Kane: Well, it already has. It can be simple or more complex. Let's say that we're looking at the effectiveness of a particular treatment, and we have two study groups. And we see that with growth our symmetry is better in one technique versus the other, then you organically move in that direction towards doing that operation.

And so, this is where the research pushes the envelope towards patient care and it translates back. We've made many changes based upon what we've learned in the AIM Center in terms of how we actually treat patients. And this is where the AIM Center has another important outcome, which is it allows us to up for the entire institution the academic profile.

I mean, academically, part of our mission is to obviously take great care of patients. The other part is to teach the next generation. And the third part is to move research forward. This area allows clinicians to bootstrap clinical information into academic production. And this is an important thing for the prestige of the institution and important for academicians who want to move their career forward. And this again cuts across not just plastic and craniofacial surgery, but across all the other disciplines that utilize the center.

Host: And one of the reasons we have you both on is that this work brings together surgeons, imaging scientists, other specialists. What does that collaboration really look like? How do these different disciplines all work together on an individual patient?

Dr. Rami Hallac: Yeah. So, what I think is really powerful here is that surgeons can connect very good clinical questions with advanced imaging tools. So now, they don't have to rely on very subjective measurements. They can actually measure the outcome and see which surgeries and outcomes work best. For us, the surgeons help us understand what really matters to them clinically, and we try to find solutions or mathematical model to answer that questions for them and the collaboration doesn't stop here.

So, part of what we do is we try to take something that starts as research study and try to bring it into the clinic so physicians can use it for their patients. A good example is a smartphone app that we are working on for facial paralysis. So, the surgeons help us define what's clinically important to measure, and our team helps develop the analysis and the imaging tools needed. And we also work with software developers, UX designers, to make this app more feasible for the patients. So for me, multidisciplinary collaboration that's how it should look like. You know, different people bring their expertise to the same problem, and moving it from an idea to something that can actually be used in the clinic.

Dr. Alex Kane: And I'll add onto that, I think, it's fair to say that every clinician working at this facility, at the Children's Hospital is very keenly aware of the problems in their own discipline, the clinicians are. But it is not uniformly expected or common that they have the ability to understand what these imaging modalities can bring and how to utilize them. I mean, it does take a certain amount of expertise to employ these techniques.

Where the magic happens in a multidisciplinary way is that they come with the questions, there's a dialogue and an interaction, and what is possible becomes understood by both parties, and that's where the synchrony begins. And you start to collect data that can be studied using these techniques, and that's where the beneficial virtuous cycle begins. And we've seen that over and over. And it becomes a very powerful leverage for clinicians who didn't otherwise have that expertise, because that expertise resides in a person like Dr. Hallac, who has all the tools at his disposal and understands in a very deep way what the clinician might only understand in a superficial way, and vice versa. The person who is uniquely an imaging scientist may not have the same level of expertise of what the problems that need to be solved medically might be.

Host: Sure. Absolutely. And so, as artificial intelligence and advanced imaging and computational modeling continue to evolve and very, very rapidly, where do you see the greatest opportunities to advance plastic and craniofacial care for children?

Dr. Alex Kane: I think that it's going to be very interesting to see what happens. I don't think I need to be the first person to say that there are limitations and risks associated with wide sharing of data with LLMs that no one is exactly sure of how they work.

Obviously, we are expected and very understandably to maintain patient privacy to meet all the compliance standards. But if you really want to advance the medicine, you need to have as much data as possible. So, there's this kind of tension between those two needs. And so, ultimately, I am optimistic that we'll be able to resolve this seeming paradox about sharing more data so that we can get better results in an iterative fashion versus putting up a wall. Right now, I think institutions are having a hard time catching up with the technology.

And I think that could be said across the entire AI universe that these same tensions exist, but I think they come to a particular higher intensity when you're talking about using images of people's faces, a young one's faces, and heads. It can be very difficult to decide where to draw the lines at this particular point. So, it's hard to predict the future, and it's becoming harder as the technology accelerates at such a fast rate.

Host: Absolutely fascinating. And you're right, I think that there is a tension and a balance that we're all trying to strike right now, and we're not sure exactly how that's landing, and it changes every day. As we come to a close, is there anything else that either of you would like to add to our discussion?

Dr. Rami Hallac: Just for the AI, what we're trying to do as well, like in the AIM Center, is to create images that can be used to train the AI that are de-identified or anonymized. So for example, we can train the models to create images or artificial images of cleft lip deformities. So instead of using patient data, we can use these AI-generated images to train our models, so this way we can protect patients' privacy.

Dr. Alex Kane: I feel like I'd be remiss if I didn't say that I think in my career it's been a unique experience here where the Children's Hospital has had the foresight and the willingness to underwrite this sort of a project.

This is not something that is remunerative per se, and it is something that I think is enlightened in terms of how the hospital is allowing us to do this work. Because it's not inexpensive to do, and I think that they share the vision, and I'm grateful for that.

Host: Yeah, I can see that completely. Dr. Kane, Dr. Hallac, thank you so much for being with us today.

Dr. Alex Kane: We appreciate you.

Dr. Rami Hallac: Thank you so much.

Host: Absolutely. And for our audience, thank you for listening to Pediatric Insights: Advances and Innovations with Children's Health, where we explore the latest in pediatric care and research. You can find more information at childrens.com. And if you found this fascinating podcast helpful, please rate, review, and share this episode, and please follow Children's Health on your social channels.