Episode#116| Event Recording: Advancing 3D Surgical Planning

Surgery is moving into a world where we can measure, simulate, and even rehearse before we ever touch a patient, but getting 3D surgical planning to feel “normal” inside a hospital is still a battle. We bring together a rare mix of voices across the ecosystem to explain what’s actually working, what’s still missing, and what it takes to scale medical 3D printing and virtual surgical planning without sacrificing safety or quality.

Event Page: https://3dheals.com/3d-surgical-planning/

Subscribe, share this with someone building a 3D program, and leave a review with the biggest blocker you see to making 3D planning routine.

Transcript

Copy Transcript Copied

3DHEALS Mission

SPEAKER_02 0:00

My name is Jenny Chen. I'm founder, CEO of 3D Hills. We've been hosting this for almost six years in virtual events. So we have three missions as a company. Number one is to educate the public about what 3D printing can do and also what it cannot do. And also now we have expanded, like this panel is reflecting, to the entire 3D technologies because you cannot function in healthcare in silo. It has to be an ecosystem. And as you can see later on from this panel, how technologies work with one another to solve problems. Okay, that's number one. Number two is to network. So even though virtual events is not as fun as in-person, as I do enjoy in-person events myself as well. Um, but virtual events actually focus more on the content, um, but also uh what people are doing. So if you're the audience, just type in where you are, who you are, and what you're looking for in this webinar so that our speakers can get to know you and also you get to know one another. If you have a social media account, also share it. We're very active on social, so we are happy to follow accounts and if you want to be followed, that is. And okay, number three is pitch TD program. So we have uh started to help startups since 2018. We have helped more than 100 startups, and we actually have a couple of startups here today in this program to create those first meeting opportunities with institutional investors, which means you do not have to co-call because we're already primed the investors that we're connected to about 3D printing. They understand what 3D printing is, they understand 3D technology, med tech, biotech, and so you will have a relatively warm intro to a first meeting. And this program is entirely free. And so if you're a startup founder interested in joining the program, contact me after the event. All right, so now we have more than one audience. I'm happy to see that. I'd like to introduce the panel. So this panel um it's it's really I feel like it's the evolution of 3D heels. You know, initially when I got interested in the 3D technologies, 3D printing was amazing. It's like the first c human computer interface where I can actually interact with images on my screen as a radiologist. But now as I understand, I also have deepened my understanding to the healthcare system and how surgeries happen and how surgeries is moving towards um many different technologies are now working together, both on the software and hardware side and maybe even distribution side, um, to advance the next generation of surgical techniques. And we also have a surgeon in the audience, uh in the panel, right? David, are you a cardiosthora surgeon? Is that correct?

SPEAKER_04 3:00

Um surgical oncologist.

SPEAKER_02 3:03

Surgical oncologist, correct. So we have we we have clinicians, actually, Mark is a radiologist as well. So we have quite a few clinicians today. So this is gonna be a very good conversation because people who used not to sit at the same table can now get to talk to one another. So without further ado, I'd like to introduce our first speaker, uh Tim Van Kouenberger. I tried very hard. I feel like I am there's some words for it, but I'm not gonna say it publicly. Um but Tim, please take it away.

A Custom Hip Dysplasia Implant Workflow

SPEAKER_02 3:38

Tim is the CEO and co-founder, oh CTO and co-founder uh of Replacia.

SPEAKER_06 3:44

Uh thank you, Jenny. Uh hey, good morning, good afternoon, good night, wherever you are. Uh so I'm Tim. Uh as Jenny said, I'm I'm one of the co-founders of Replazia. We're a metic company based in Belgium. Um, and we are focused on uh the hip preservation field. Uh so what I wanted to do in the uh that's in the next 15 minutes uh is explain to you how we set up a end-to-end workflow to bring an orthopedic device, custom-made orthopedic device manufacture with 3D printing, bring that uh to the patient, uh, and also share some of the lessons learned. Um, maybe to set the stage a little bit uh about hip preservation, why is this such an important field? Um, this is this is a story of Amanda. So this is uh a patient story uh that's also available on the International Hip Dysplasia Institute. And so um Amanda was a young adult, a young girl, uh let's say age 17, 18, was doing cross-country competitive running. Um, and then all of a sudden she started having complaints. She started having hip pain during and after races. Uh, she got an MRI scan, went to see doctors, and then she was diagnosed with uh um what is called hip impingement, which is um an excess bone either on the femoral side or on the uh the pelvic side, on the acetabular side, which causes uh painful contact and limited mobility. And she was also diagnosed with uh labral tears, the the sort of the seal um that that seals your your hip joint. She had some tears in there as well. So she underwent uh a hip artroscopy, um, which is a typical standard treatment for for chem lesions or for impingement, um, that gave temporary relief. Then she underwent a second surgery uh that again gave temporary relief, but then she went further downhill. Uh her complaints got worse and worse. She was bound to her bed and to a wheelchair. She had to stop her studies of becoming a nurse. And then after some uh more examinations, finally she got the diagnosis that she had hip dysplasia, which is an underdeveloped hip cup, so a shallow hip cup, uh, which causes early wear of the cartilage if left untreated, uh, also pain and so on. Um, there are hip reservation treatments for this that I will cover in my talk. Um, however, for Amanda, unfortunately, her uh osteoarthritis was had evolved too far uh at that time, by the time she got the right diagnosis. So the only solution left for her was to get a total hypostasis on both sides at age 22, which is far from ideal, of course, because taking into account that the lifespan for the lifetime of these implants is let's say 20 years, she will need uh at least a few revisions throughout her lifetime. Um so I think this is a nice illustration of how important this field is and why it's so important to get it right first time, because these are young adults. And so hip dysplasia and impingement uh actually uh are more prevalent than than one would think. It's about 10 to 15 percent of the population suffering from either dysplasia or impingement. Uh, it's the most common cause of hip pain in adults uh under 40, and as you saw in the story of Amanda, uh dysplasia can be a leading cause for osteoarthritis if left untreated. And these are patients in their 20s and 30s. So if possible, a total hip replacement should be avoided at all costs. Um, so what are some of the current uh treatment options, surgical treatment options for hip dysplasia? On the one hand, you have what is called a shelf acetabloplasty. And so keep in mind with hip dysplasia, the hip cup is underdeveloped and there's instability of the hip. So with the shelf acetabloplasty, what they do is they take a bone graph from the iliac crest, make a slot above the hip joint in the ilium, and then place a bone wedge, this bone wedge inside the slot as a sort of a buttress that helps in stabilizing the hip joint and also then distributing the loads. Um, this has pretty good outcome actually when done correctly. Um, it has some surgical challenges in the sense that it's very important to place this uh this um this uh bone wedge in the right position, not too high, but also not too low. Um, and it has been abandoned because of this this uh challenging surgical technique and and the importance of uh precise uh positioning. Has it been abandoned in favor of uh nowadays uh what is called the periacetabular osteotomy? And there what they do is they make an osteotomy, they release the acetabulum from the rest of the pelvis, re-order reorient it in a more favorable position, and then fix this with very long screws. Now it's pretty obvious that this has is a very uh very evasive uh surgery. Um it also has a very long recovery time. We're talking about six to twelve months even, and has quite a high complication rate, but still it has good outcome, but it does have some limitations. So the surgical techniques that are available do have a pretty good outcome but limitations. So we made it our mission to um to develop a new, a novel treatment for hip dysplasia and young adults. Uh, so we wanted to develop a minimally invasive procedure that had a more predictable outcome, uh faster recovery, and also reduce some of the complication risks like resorption of the graft that we've seen uh sometimes with with these uh shelf acetabloplasties and also avoid uh uh impingement, creating impingement problems. Um so what we did was we reused the concept of this shelf acetabloplasty because it it does have a good outcome and a good success rate. Um but instead of using a bone graft, we decided to use a patient-specific implant, which is designed based on CT and then manufactured using metal 3D printing uh in titanium. And so uh as I said, we we start from this concept and then we have developed this implant that you see here on this bone model. So this is a titanium implant with the the lower part, is actually the the active functional part of the implant, if you will, that sits on top of the joint capsule. It does not go inside the joint, it sits on top of the joint capsule, but it provides additional coverage uh of the femoral head, uh additional stabilization of the joint, and it distributes the loads better, reducing symptoms and and slowing down the development of osteoarthritis. Um now, when we were developing this uh this product, um we noticed that we we needed to develop a new software tool, better software tool, because we were developing a patient-specific uh device. Um so we needed to be able to measure with high precision very specific anatomical parameters of this of these patients' uh hip joint based on CT images. Um, we also needed to identify where the acetabular defect is, so where the acetabular in fact is is too small, and where we needed to provide or where the implant needed to provide that that additional coverage. Um, and also we need to make sure that the that the implant nicely fits on the patient's anatomy uh and that the results ephemeral under coverage. Um, and we wanted to be able to verify the implant geometry uh once placed inside the patient against the range of motion to make sure that there's no limitation of the range of motion of the uh uh of the hip joint due to the presence of this implant. Um so we developed our own software. Uh, this software uh allows us to make a very detailed uh anatomical model. This is not new, obviously, this has been around for quite a while to make a 3D model of uh CT images. But what our software does is to really uh make very specific measurements. For example, the blue area that you see here is the part of the femoral head coverage, uh or the part of the femoral head that is in fact covered by the acetabulum. And then the middle graph here you see again that blue line that that maps in a polar graph what that coverage looks like. And then we can use this as well to decide where this acetabulum needs or where that coverage needs to be extended. So that then serves as input for the implant design. And then also with the software we can do motion simulation and see if there's uh early contact anticipated between the implant and bone, and then we can iterate and and adapt the uh implant design uh as needed. Um so then we showed this these measurements to surgeons. They were very positive, very enthusiastic about the measurements that came out of this, um, which led us to the decision to develop this into a separate product or a service in this case. So, using that software, we now offer a service, which we call a hip studio analysis service. So we use the software to generate a report, very detailed report, um, to support diagnosis and treatment planning of hip preservation, both impingement and dysplasia. It measures a wide range of anatomical measurements that are also relevant for the 3D shelf implant. It allows a kinematic simulation, impingement detection. Uh, we also include healthy reference data for easy interpretation. Uh, the software itself has AI and machine learning uh built in, for example, for image segmentation and the detection of impingement. Uh, and there's all sorts of uh measurements that we do. This clock graph that you've seen, uh, the femoral antiversion. This here is uh a measurement of uh the impingement to detect where the access bone is on the femoral neck in this case, and then you have the range of motion, just some examples of some measurements that are included. Um so we have the soft, we have the implant. Now I want to quickly run through the whole the whole workflow. Uh, so what does that end-to-end workflow look like? Um, as I said, we start from CT images. Um, we have a specific CT scan protocol that needs to be followed, uh, which uh requests that the full pelvis is scanned, but also the knees should be scanned because you need that information, uh, for example, to measure the femoral anteversion. Um, then those DICOM images uh are uploaded to a secure cloud platform. We download the images, then uh we make uh accurate 3D models using off-the-shelf software. Uh, and then we use our own software to do all those analyses that I just mentioned, and then there's a PDF report that is generated and that's then provided to the surgeon and helps the surgeon in setting the right diagnosis, and as you've seen in the case with Amanda, um the goal is obviously to give all the information that the surgeon needs to set the right diagnosis right uh from the first time, and then the surgeon can decide with that information what the best treatment, what the best course of action is. Um, so you can choose for conventional uh treatment, for example, physiotherapy, or you can choose for surgery, and if he then wants to use the 3D shelf implant, then we could we use that same CT data to design the implant to see where that coverage defect is and what that implant needs to look like. Um that is then reviewed and approved by the surgeon. Um, once the design is approved, it's manufactured using metal 3D printing. Um, there's also a guide that is manufactured uh and provided with the implant to make sure that the implant is placed in exactly the planned position. Uh, there's an atomical model that we print in nylon using SLS and then also an instrument to help in hammering the implant uh into the uh into the all the way flush to the bone. Um these this package is assembled together with all the paperwork, the surgical protocol uh and whatnot, and that is then shipped to the hospital um to allow the surgeon to uh to treat the patient. Uh so that's the workflow.

Clinical Proof And Regulatory Reality

SPEAKER_06 15:30

Um, in in the last uh section of my part uh of my talk, I would like to um uh mention some of the or highlight some of the challenges that that we've seen as a medtech startup. Um, one challenge is is the need for clinical validation. So we need to demonstrate that this concept actually works in humans. Um the 3D shelf implant is a novel concept, there's no predicate device. Um, there's an extensive canine study that was done at the University of Utrecht uh with excellent results. And here you see uh cross-section of uh of one of the uh of one of the dogs that was uh treated uh uh and then um once the dog passed away they did a cross-section. Here you see the implant here in gray, you see the screw that was used to fix the implant, you see the acetablar bone. Um, and here is the acetabler roof, and here very interesting. You see the the um the the uh the joint capsule, and what you see is that this has become quite thick and fibrous cartilage-like tissue. And this is very interesting and very important because that explains why in dogs it they see such good results. Um, so there's the it's also based on a uh uh on an existing surgical technique. Uh, in I mentioned the shelf as tabloplasty. So there's there's good indication that this might work when we replace the graph with with a metal implant. But still, we need to demonstrate that it works uh in humans. Uh and this is why we launched the first in-human clinical trial that's now ongoing. Uh, it's 10 patients in a single center study in the Netherlands, and the first patients uh were in fact treated uh successfully. Um, the uh the second challenge is institute readiness. Uh, when you introduce a patient-specific device, whether it's an instrument uh or an implant, uh, it's not just about how to use that specific device. In this case, it's not just about teaching the surgeon and the OR stuff how to uh how to place the implant, but there's much more that that comes into play. There's uh the CT scan that needs to be taken with the correct scan protocol. They need to upload the DICOM images to a portal. So those are uh that's also a deviation from their standard way of working. Um, there's a lead time for implant manufacturing. Um this is elective surgery, but still uh we need a couple of weeks to to design the implant, finalize the design, and then manufacture the parts. Um and that needs to align with the surgical scheduling, of course. And then there's also the hospital logistics. Um so this is a patient-specific device. It's made to order, of course, it's not an off-the-shelf device. So it needs to be ordered, then once it's received, it needs to be matched to the right patient, uh, obviously. And then the hospital also needs to order the right screws and make sure that there's some additional instruments that are needed to place the implant or are also ordered. So these are also some things to take into account. So it's it's not just about uh a different surgical technique, uh, it's really an institutional change. And then the last is, and this is probably not new to the audience, there's a regulatory clearance. Um it's a patient-specific device uh combined with 3D printing. So we had to set up an ISO 13485 certified quality management system uh for both software and hardware. Um as far for the device classification, uh, under MDR, it's a class 2B device, uh, but under the custom-made device exemption because it's a really a custom made patient-specific device. Um, for the FDA, uh there's no predicate device, so that will be a de novo pathway. Um, and and obviously collecting clinical evidence and uh the clinical study and then going to the regulatory approval that really drives your your go-to-market timeline more than anything else. So, to conclude, uh, a couple of lessons that we learned from this end-to-end patient-specific implant workflow. Um, this is maybe preaching to the choir, but always start from the patient and start from a real clinical need. As engineers, and I'm I'm an engineer myself, it's very tempting to look at all this cool technology, AI and 3D printing and whatnot. Um, but it's important to first start from a real clinical need that will really impact your and then and improve your success rate uh more than anything else. So first look at the clinical need and then start building the product and then see which technology is the most suited. Um, also building a patient-specific solution means building the full stack. It's not just the product, it's a quality measurement system. In our case, we need to develop a software. Um, there's also the full process validation for the 3D printing. We outsource the production, so that's done by our suppliers, but still, all the software tools that we do for processing the images, designing the implants need to be validated. And then lastly, the hardest problem is not always the technical uh problems, like the clinical validation, collecting good, solid clinical data, um, making sure the institutes are ready to use your product, and then that regulatory clearance can be even a bigger, bigger uh uh challenge. Um, so that's what I had. Um, thank you very much for your attention. If you have any questions, and I'm more than happy to answer those, of course.

Scaling Barriers And Cost Questions

SPEAKER_02 20:45

Okay, so um Tim, that was an excellent presentation. I just realized the QA box is gone from Zoom and I couldn't find it because I was telling people if they have questions. So they they change the interface almost on a daily basis on me. So if you have a question, maybe just put it in the chat right now. If you do see the QA box, maybe put a question in there just so I can see it. I have no idea where it is. Um, I have some quick question just to just to kind of start the conversation. Is one is are the surgeons already doing the 3D planning now without even without your software? Are they like they're they're already thinking in 3D or are they still using the films to plan these surgeries?

SPEAKER_06 21:25

Well, there's um the surgeons that we talk to, most of them use uh CT uh for their more complex cases. Okay. For very, very pronounced uh dysplasia case, they might stick with 2D uh uh uh AP radiographs. But for complex cases, uh most surgeons will use a CT and then at least they have that 3D information. What we do see is with um they're very much still uh and not all of them, but some are still thinking in 2D, or at least they're used to seeing 2D data. They're measuring 2D angles from 2D radiobacters. After a while, in fact, the CT has that full 3D information encapsulated. So it's like alpha angle, uh lateral CE angle, those are traditional measurements done on a 2D X-ray. And that's what they typically expect even if they have, or what they look at even if they have 3D uh information. And you see that changing, but that's still an ongoing process.

SPEAKER_02 22:22

Probably because that's probably the largest kind of data we have in history collected from this. So we have a reference. I'm hoping people are now having some kind of reference in 3D. Do you guys use like a population statistics to generate your prediction or your analysis?

SPEAKER_06 22:41

Yeah, so what we do is we um for all the measurements that we do, we have normal data for a healthy hip joint for each and every measurement. And that's provided as a reference. Uh and that that will help in in building that that 3D thinking, if you will. And uh because it's you're absolutely right. They are um still thinking in 2D because most of the reference data that was generated historically was on 2D or X-rays. Um but you see that that will my expectation at least at least is that that will change over the over the coming years as more and more 3D data will become available. There will be 3D standards that become available for all of these measurements.

SPEAKER_02 23:24

Oh, somebody put a question. That's awesome. I found it. Thank you, anonymous attendee. Uh it says, what do you see as the biggest barrier to scaling patient-specific CT-based implant solution to into routine clinical workflows, and how is Replasia addressing it?

SPEAKER_06 23:44

Um I think one is um being able to provide this in a cost-efficient way, um, because these are made to measure, and every implant needs to be designed uh for each specific uh patient. And that's one thing that we try to solve by automating and scripting, automating the design process and using scripting uh to automate that and to reduce that that manual labor, that manual time that's needed to actually design the implants. Uh and that then ties into the cost, of course. I think that's another challenge. This I think many people will agree that that patient-specific um, like in orthopedics, patient-specific devices have have definitely certain benefits over off-the-shelf standard uh devices, uh, but there's a cost associated with that. Um try to keep that cost uh low so that you can place it in the market at a competitive price, that it becomes a valid, a value uh let's say a valid uh and a viable alternative to the off-the-shelf uh solutions. Um and then the other thing is reimbursement, of course. Uh we're still in a very early stage because we're still only now doing the clinical study. But yeah, we know both in Europe and US reimbursement is important, but but it's also a challenge. How we're gonna tackle that, I think that that's still that's still uh something to tackle once we really go into the market.

SPEAKER_02 25:15

Yeah. Um real quick question is because I saw the box assembled for uh for the replaceia device. You have in fact two and not just one implant in the box. Um why do you have two?

SPEAKER_06 25:28

Oh, the second one is the guide, in fact. We're at the moment we're we're making the guide also in titanium. Um yeah, so that's uh for a couple of reasons. One is then we know that it's gonna be strong enough. We don't need to glue in metal bushings to guide the drill because the guide itself is is completely made out of the yeah, it's completely made out of uh of metal. Um so we have the implant in titanium, then the guide as well in titanium. We might uh um change that to an SLS guide in the future, but for now uh we're using titanium guides. And then there's an anatomical model that we provide, and then uh an instrument that fits on onto the lateral side of the implant to really hammer the implant all the way to until it's flush with the bone. So it's one implant, one guide.

SPEAKER_02 26:14

Got it. Okay, we have a question from um Arpin. Uh I'm gonna shorten it basically outside of um uh you you you mentioned a lot of challenges in adoption, hardware, software, QMS, clinical. Which one is your biggest barrier, you think? Of all the pain points, what did what is your biggest pain point?

SPEAKER_06 26:38

Um but that's that there's there's there's yeah, I'm not sure if there's one one that that really sticks out. Um I think getting the the regulatory approval for the clinical study uh was quite challenging because then you present everything to an ethical committee and they review your file and then they come back with questions and they request more data, so you need to do additional testing. So that has taken us quite quite a while to get that uh that approval from the ethical committee. Yeah. Yeah, I think I think that's probably one of the bigger hurdles that we that we encountered to get that approval.

SPEAKER_02 27:14

Um are you incorporating any automation right now into your software, or is that just a future goal?

SPEAKER_06 27:23

Uh no, no, there's there's uh automation built in it. So the the the way the software runs is semi-automatic. So there's still some manual interaction required. Some process steps we have automated already, and we want to further automate that also using AI and machine learning and so on. But that's work in progress.

SPEAKER_02 27:40

Okay, great. We have one final question, but um I think we're running out of time a little bit. I want to keep it going. We have a lot of speakers today, so there is a question in the chat box. If you were to answer that directly, Tim, that would be great. Uh, I'm gonna move on to our and thank you for the presentation and QA. That was very insightful. Um, next I want to introduce our second speaker. Let me see, Beatrice. Okay, great, is uh Beatrice Gonzalez. You're the global market manager for Materialize. Also, I want to mention that I want to thank Materialize for sponsoring us for this event. Without our sponsors, these events will be a lot more challenge to organize.

SPEAKER_08 28:20

Well, thank

Three Shifts Shaping 3D Planning

SPEAKER_08 28:21

you very much, Jenny. Thank you for organizing this, for giving me the opportunity to be here today. As Jenny said, my name is Bea Domínguez González. I am the global market manager for software in hospitals and universities at Materialize. And over the next 15 minutes, I want to share with you what I think are the three shifts that uh will matter for the future of 3D surgical planning. Some of what I'm going to talk today comes from peer-reviewed clinical evidence, and some of that comes from what we see every day across a network of more than 500 hospitals that are using 3D planning in routine care with our software. So, my goal today is to give you my view, my clear view on where is 3D surgical planning in 2026 and what still needs to happen for it to become routine rather than exceptional. Uh, I like to begin with uh with a short story, uh, not because it's extraordinary, but it is because it demonstrates what becomes possible uh when the uncertainty is replaced by predictability. In June 2021, Aaron was working on a high voltage power line when his face uh accidentally touched the wire. That was 7,200 balls passed through his body. Uh he survived, but the injuries uh resulted in the loss of his left eye, uh large portion of the nose and mouth, significant midfacial tissue, and his dominant arm. As you can imagine, even daily basic functions became a real challenge for him. Uh he went for two years in multiple reconstructive procedures and extensive rehabilitation, and back then uh he reached the limit of what medicine had done until that point. Uh however, the clinical team at NYU Langon uh thought that they could do something more for Aaron, and they uh they believed what was impossible back then. What if we transplant not only the face but also a human eye? Whole eye transplantation had never succeeded. Uh, the constraints were very well known: restoring the blood flow, maintaining the structural stability, and preventing the rejection. And each of those challenges had been limiting progress for decades. However, in May 2023, more than 140 surgeons, nurses, and specialists entered the operating rooms in New York City to achieve what back then had never been done. For 21 hours, there were two surgical teams that worked simultaneously on the donor and on the recipient, attending the first combined partial, face, and whole eye transplant in human history. But of course, the eye could not just be transplanted independently. It had to be attached together with all the surrounded bone and tissue and implanted into the recipient with millimeter level accuracy. In preparation for the surgery, materialized clinical engineers worked very closely with the surgical team, and using high-resolution CT imaging, they've created a virtual 3D anatomical model that allowed the team to rehearse the procedure and align on clinical on critical decisions before they enter the OR. And based on that planning, materialized designed and 3D printed the patient-specific cutting guides, both for donor and receiving. Once the suitable donor was identified, the instruments were manufactured in under 24 hours and delivered in time for surgery. It's important to mention that, of course, the planning technology did not replace the surgical expertise but amplified it. It enabled the predictability, alignment, and precision that made this surgery possible. Post-top, the patient uh regained critical functions, uh, including nasal breathing, oral intake of solid food, and fascial structural integrity. You can imagine how life-changing all of those are. The transplantant eye maintained blood flow, normal pressure, and stability as well. And one year follow-up, um, there were no signs of rejection, and retinal cell survival was confirmed. Um, so it was uh a real scientific uh milestone. Now, the case of Arrow represents the outer edge of what is possible today. Uh, but the tools behind it, 3D anatomical model, virtual planning, patient-specific instruments, are already common in many hospitals. So the question no longer is whether the technology works. The real question is how we can scale it, how we can standardize it, make predictability routine rather than exceptional, how we make it to reach more patients. Uh so if the goal is to move from these exceptional cases to routine cases, what actually has to change? Uh, from what we see across clinical evidence and day-to-day hospital workflow, the change isn't driven by a single breakthrough technology, it's driven by a set of smaller shifts in how 3D planning is understood, used, and integrated into care into each hospital. I want to focus on three of those shifts today because uh together I believe they explain how 3D surgical planning is moving from something impressive, from wow stories to something dependable, for something that we really can count on on the day-to-day clinical practice. The first one is what I'm calling beyond the print. Uh, there is a misconception that the goal of 3D in healthcare is always the printed model. And well, it isn't, right? Like what we see is that although the 3D print is essential in many clinical applications, in order, most of the clinical value sits in the digital plan itself. Uh, the second uh is on surface modeling, not just volume rendering, becoming the standard for surgical planning. There is a clear distinction, a technical distinction that has very real clinical consequences, and I'm going to talk about it a bit later. Uh the third one uh is the honest part, is also what Tim was referring to earlier, the gap is might not be anymore the technology, but uh reinforcement operations, evidence, um it is still holding us back. Uh it is slower, it's less visible work, but there it is where uh I believe the progress now depends on. So now coming back on the 3D uh surface modeling, um most of us are very familiar with 2D images stacks and volumetric renderings. They are excellent tools for diagnosis, they were designed for diagnosis. Uh, but where they begin to fall short is in the surgical planning. Uh, a 2D slide gives you a cross-section, and volume rendering gives you a three-dimensional visualization of a picture. It remains a picture. There are no defined objects in it. Uh, for example, in the case of a tumor and the vessels and the surrounding tissue will be rendered at one fuse image based on the box cell intensity. Uh, so you can see something there, but uh the software has not identified what is what. And uh in soft tissue where, for example, a tumor and a vessel next to it might share similar intensities, they can blur into the same region, and you might not be able to reliably separate them and measure between them or plan around with them. Now, with surface uh modeling, something different happens. It segments each anatomical structure, uh bone, uh basin tumor, soft tissue, into a discrete 3D object with explicit geometry. You can rotate it, you can hide it, measure it, isolate it, and test decisions against it. Uh, that's uh unlocked uh three things. Uh it is interactive at the object level, so not just a viewpoint that you can rotate, but the anatomy you can take it apart, you can reassemble it. It is also measurable, uh, distances, volumes, uh, the margins are computable, uh, which matters when you are choosing a clumping strategy or a screw trajectory. And it is simulatable. You can rehearse the procedure, you can 3D print, you can 3D print as we have seen before, uh tools for the surgery, and you can also bring it into AR before you go into the UR. Um so this is a shift that surface modeling enables from viewing a tool to a planning tool. I want to bring just uh uh one example of why that uh distinction matters.

Why Surface Models Beat Renderings

SPEAKER_08 37:12

I just went one before I shoot. Yes, I just want to bring one example to you on why that distinction matters. Here I'm sharing a prospective study from IRCCS Bologna published in Frontiers in Oncology in 2022. Uh in this study, 195 patients uh were scheduled for partial nephrectomy. Um and one group received a standard 2D CT review with the 3D surface model, and the control group received the 2D CT alone. Uh the primary outcome study was trifecta, so no positive surgical margin, no major complications, and an EGFR drop of less than 30% post-op, uh, which it means clear cancer surgery, no major complications, and the kidney is still working. Uh basically the outcome that every urologic surgeon is uh looking for. In the 3D plant group, 80% of the patients achieved the trifecta. In the control group, that dropped to 63%. Those were the same surgeons, the same CT scanner, the same patient population. The difference was the presence of a patient-specific surface model of the kidney, the tumor, the vasculature, the collecting system in front of the surgeons before they arrived to the OR. It was the same CT scan, but a better surgical plan and better outcome. In this case, the clinical decision changed because of the virtual uh model. Uh, the print is still required when surgeons need a tactile feedback or a sterilizable guide for the OR. Um, however, across different disciplines, urology, oncology, even CMF and orthopedics, depending on the application, the plant delivers uh the value. And sometimes virtual only it's enough. Sometimes, of course, in uh orthopedic, CMF, and even other disciplines, the printed is uh preferred. Um so CT and MRI are inherently pretty modalities and images, imaging specialists are already in the day-to-day interpreting volumetric data. When the surface modeling lives in-house, when this service is taken inside the hospital, the same scan that establishes a diagnosis also can become the surgical plan. There is no second exam, there is no external handoff, there is no parallel workflow. This brings several advantages: control over turnaround and quality, alignment with existing imaging workflows, and subspecially expertise that stays within the hospital. What matters the most is without doubt the communication. When the surgeons and the imaging specialists work from the same patient-specific 3D model, they share a visual language. So the anatomical variance, the artifacts and the constraints surface during the planning rather than during the surgery. So surface modeling, of course, will not replace diagnostic interpretation, it just extends it. To make 3D planning scalable inside a hospital, the tools uh matter. What we see work in practice is not a collection of isolated tools, but a single planning environment that connects segmentation, measurement, simulation, and when needed, also the design of patient-specific instruments. Um materialized that environment is MIMIX, and the important point though is that uh is that connection is the connection that the platform enables. In practice, that means integrated case management for quality and traceability, automation and AI to save time, more accessible 3D interaction through AR or XR so surgeons can engage directly with the plan. This supports an important mindset shift. A connected planning environment keeps the focus on the clinical decision while quietly supporting scale, consistency, and collaboration across specialties. So, where are we today? On the ready side, uh surface modeling accuracy is high, even for complex anatomy. Software platforms like Materialized Mimics are mature, and workflows are proven across more than 500 hospitals globally. Where we are catching up is in everything around the technology. Operational integration takes time, the evidence is still being built, indication by indication, reimbursement frameworks are evolving, and adoption spreads laterally, surgeon to surgeon, which is very effective but it's slow. So this is no longer a technology problem, it's a system problem. Now, if you are considering to invest in a 3D lab, my advice is straightforward. Invest in the workflow. Start with surgical services where complexity is the highest and the value is immediately visible. We often see it in cardiology, oncology, CMF, complex orthopedics. Here is where surgeons often see the difference immediately, and that builds internal momentum. The second uh advice would be to standardize segmentation protocols for defined indications. I had it again. Sorry for that. Um yeah, so having a clear protocol per integration is what makes the work reproducible. And on the third step is to align the turnaround with the surgical scheduling. If the 3D planning had the 3D planning has to fit the OR calendar, not the other way around, because obviously if we are not having the plan ready for when the surgery is happening, uh then it basically would get used. Um scale in our uh what we see over uh 500 hospitals around the world is that uh the scale comes when doing this well in two or three indications, not everything at once. It's very tempting to start with a lot of illusion and wanting to do everything at once, but we advise to start small, prove the workflow, and expand uh from there. One signal that the adoption is real is reinforcement. And uh there is several initiatives worldwide ongoing to secure it. I'm bringing in here one example from the US. So in the US, in September 2025, the AMA approved three sets of new category three CPT codes for digital workflows. Those are for digital 3D model, digital simulation, and computational analysis. These codes are going to become effective in July 2026, and they recognize surface modeling and virtual planning as a distinct clinical activity independent of 3D printing. Uh the 3D printing ones uh still remain and they they are still valid as well. The category three codes are tracking codes. They allow utilization data to be collected and help to build the case for permanent reimbursement. The practical implication is very simple. From July, the work that many of you are already doing now has a code, has a recognition. It is also worth noting that a dedicated registry for medical 3D planning will be launched soon under the leadership of Dr. Fran Twibic and colleagues. It is designed to capture the clinical context, the use of 3D, and the downstream outcomes. Once it is live, so stay tuned. The field will finally have a structured mechanism for generating the evidence that is needed to support the future reimbursement. So please document it, use it consistently. So the data that you are generating will shape what the reimbursement will look like in the years ahead. What remains is adoption, integration, and the steady accumulation of evidence that will turn these exceptional cases into the standard of practice. What made the eye transplant possible is the same thing that makes everyday surgery safer. Seeing the anatomy clearly planning before entering the operating room in a 3D complete uh view. And with this, uh I want to thank you. If you want to keep the conversation, of course, we are going to have some time now, but also just uh um you can uh we can talk through LinkedIn or any other way. Thank you very much.

SPEAKER_02 45:47

Thank you,

New CPT Codes And What They Mean

SPEAKER_02 45:48

Beatrice, and also everybody, um, make sure that you enter your social media account or LinkedIn link if you want to connect with anyone on this platform. Uh it's very good opportunity to do so. Uh Beatrice, I have a question about the CPT code because uh that's news to me. Um CPT code three is still experimental, you still need to collect clinical data, is that right? Just want to make sure. Okay. And then we have like pre-existing CPT code three for some other 3D printing, is that correct? And now just these are just new ones.

SPEAKER_08 46:18

Yes. So the previous ones, they were associated to the printing side. So they are for printed anatomical models and printed guides. These ones are for the virtual side of things. So they have we have these three levels of complexity where the first one is the virtual model, so not the printing. If it's the printing, it will go into the other one. Uh one example of the surgical planning would be uh, for example, a patch planning for a congenital heart disease. Uh, and then the third level is on CFD and FEA. So this type of computational analysis where are often used, for example, uh for placement of implants can be used for an FEA for a hip, for example, but also use biology a lot in in fluid dynamics.

SPEAKER_02 47:07

I'm surprised that that's not a code yet, because that technology has been around for a while. Um okay. Now, can people submit their cases to some registry like 3D printing so they can is there a link for that? Because um I think some people here could be interested.

SPEAKER_08 47:24

Yes, of course. It's not yet available, so we didn't include it. Uh so any time soon. I I didn't uh yeah, keep tuned because it's it's going to be it's going to be very, very soon out there.

SPEAKER_02 47:39

Um actually have I feel like this is gonna be actually faster than the 3D printing part because the virtual data has a lot more users, probably. Uh I'm I'm just guessing. I have no idea.

SPEAKER_08 47:51

Um, and I think also the technology now is going faster, right? So we have also automations. The when the 3D printed started, the industry was a bit also more, everything was more manual. And so I will not be surprised to see that now everything goes faster just because in general the market is way more mature and uh it's way easier to use now and to have uh software usage that it was 15 years ago.

SPEAKER_02 48:16

Yeah, absolutely. Okay, we have a question from Anna Calfey. Um thanks so much for the presentation. He's uh do you think this kind of platform, especially the segmentation tasks, is more suitable for an engineer or for a clinician? Who's gonna do it? Do you actually need a person to do it at the end?

SPEAKER_08 48:36

Yes. Uh so um we see still uh most of the times it's still an engineer taking care of it. Um I think in the many years ago it was more a matter of the complexity of the software. I think today it's more a matter of who has the time to do it. Uh we see especially more in Europe, a lot of clinicians using it themselves. Of course, we are working a lot on automation, on AI to make it easier and easier. So uh many, many clinicians are using it, but most of all of the users are still going to be engineers.

SPEAKER_02 49:12

Okay. Well, that is uh definitely a trend that we need to monitor closely to see exactly how this is gonna evolve and what role biomedical engineer is gonna play in the industry, as well as the clinician, obviously, who's gonna do the surgery. And radiologists. I mean, I haven't nobody has asked actually asked me to plan anything so far. Um and I work with tertiary hospitals. So, all right. Well, thank you, Beatrice. Let me see. Uh code for oversight segment. Uh yes, there is the code for physician oversight right now, I think is in radiology, and not sure they're completely separate from other 3D coding.

SPEAKER_04 49:54

So I'm I am certainly not the expert on it. I I remember that there's a code for doing the segmentation, and then there's another code for the physician sort of overseeing it.

SPEAKER_02 50:05

And oh oh no, I don't think there is one, but I could be wrong, Beatrice. You have a comment about that?

SPEAKER_08 50:11

I'm I'm actually not an expert in the coding. I want to be very careful of the information I'm giving. I know about the C800 uh one, um, but I will I I don't think it covers exactly the same as the virtual models.

SPEAKER_02 50:26

Um I don't think there is. I don't think there is a separate code.

SPEAKER_04 50:32

My point that that when when we get to the point of accepting all of this as a part of pre-surgical planning, there's going to have to be a step at which a doctor says, yes, I agree.

SPEAKER_02 50:44

Yeah, there is. Yeah, there is.

SPEAKER_04 50:45

Right. Yeah. Is there a code to pay the physician?

SPEAKER_02 50:50

I don't think so. I don't think so, David, for now, but we'll see. We'll we'll closely monitor how this is gonna pan out. But I I mean the audience, you can tell me if if I'm wrong, but I don't think there's a separate code for physician, just so you know. All right. Okay, let's move on because we have a full panel, and we're gonna move on to our next speaker, Dr. Mark Tan. Um, and Mark, uh kudos to you. Every time I see you, I'm like, I wish I can be Mark. Um you're right now how what time is it in Singapore at the moment?

SPEAKER_07 51:25

Oh, it's uh 12 midnight, but it's still okay.

SPEAKER_02 51:29

Well, you look really energized. I don't know how you can stay energized at this time. Uh so thank you so much for sharing your experience in Singapore and also all over the world. You're traveling all over the world all the time. I can see you everywhere. Um, so please share your screen and look forward to your presentation.

SPEAKER_07 51:49

Okay. Uh, does it show in full screen?

SPEAKER_02 51:53

No, you have to go to the presentation mode again.

SPEAKER_07 51:56

Okay, let me do that.

SPEAKER_02 51:58

It's kind of too small. Yeah.

SPEAKER_05 52:00

Okay, let me let me try that again. I'm sorry. Is it showing yet? Let me just um Okay. Is that showing? No, is it showing up?

SPEAKER_02 52:29

Yeah, okay, yeah, it's showing up now.

SPEAKER_07 52:31

Okay. Okay. Okay, great. Yeah. Okay, nice to meet everyone. My name is Mark.

Building A Hospital 3D Center

SPEAKER_07 52:36

I'm a radiologist and the clinical lead of the um 3D printing center at the Singapore General Hospital. I'll be giving a presentation on patient-specific uh VSP and clinical 3D printing, um, what kind of makes it happen and some of the applications. I'm going to talk about this from the hospital's point of view. So I I think um, yeah, for this audience, I think I need not go too much into detail about what is imaging and what's segmentation, what's modeling, and what's CAD. But basically uh in VSP we put this all together. So this is uh uh these are the the inferior epigastric arteries uh here that are um are mapped out in the setting of a diet uh when a when a inferior epigastric flat needs to be um used to reconstruct the breast in breast cancer. Uh and what uh is done is that the surgeon really needs to know where these vessels are to avoid them doing the surgery. So uh what's done nowadays is that we're able to actually uh come up with um with models to to uh to to map out the the vessels to make sure that the surgeon knows where they are. Um and then what happens is the second step is clinical 3D printing. So uh that's uh really the functions of actually um taking the model and making it to physical form and actually using it and turning it to device and putting it on the patient, like we see here. So um getting those definitions out of the way, um, why do hospitals do um VSP 3DP? Four main reasons. Uh one is treatment efficacy. I think it's a lot of creativity uh when we work with the doctors, when you work uh in doctors and clinicians, you always we always try to do our best for the patients. So, like uh, for example, in this patient with uh where there's uh uh where there's uh um uh an osteocondroma and there are vessels around it, the surgeon wants to be able to cut through the base. Or say, like um in a patient um where the the where the patient has maybe had a had a fracture and wants to uh mobilize the hand in a certain way, but it's got like the A B graph, uh they want a custom device to make that happen. Um so that there's really a lot of um move nowadays to kind of kind of put your creativity into the devices that you build really from the hospital's point of view and from the clinician's point of view. Um the other thing that clinical 3D printing does is that it saves time. So um for some of the operations, we're actually able to decrease the operating time by a substantial amount, make surgery a lot safer by being able to kind of accurately know where you cut. Um the third thing is that there's a lot of new materials coming on, uh, a lot of new medical materials, uh polycapolectone, uh, bioresorbables, um, so on and so forth. And the hospital needs a point of a port to uh to incorporate these technologies in um as you move into things like tissue printing, organ printing, um, and and um the hospital needs to find a kind of a port to make that happen. And and the and the last is that uh we're always concerned in our day about uh kind of resiliency and supply chain resiliency. So like during the time of COVID, um hospitals will find their way to print their own temple bones, uh, their own surgical models like we were to kind of um when they burned enough cannabis. Uh and also um in the recent conflicts, uh there had been some interruptions in, say, medical titanium, uh, and um hospitals had to kind of um redirect the flows, and and there was some time before um surgeries could actually take place. So, kind of having gone through that shot, uh hospitals are now looking into actually quite seriously into how to secure uh supply chains for uh key medical uh devices, and that's kind of where 3D printing actually enters the whole conversation on this. Um but that that's not enough for hospitals. I think there are three main things that um they need to prove. One is that it must demonstrate clinical efficacy, and I think now we're in a space where uh 3D printing was um became more prominent in about 2015. In 2020, I I think a lot of trials came out, and then now what we're looking at is uh we're really looking at the meta-analysis of the trials that are showing the output in the areas that matter to clinicians, um, operative time, blood loss, um, accuracy, so on and so forth. I think the evidence is slowly coming out in many different disciplines. And um, and but that's still not enough for the hospital. We have to take that one step further and translate it to healthcare, uh, the whole value to healthcare. So, for example, for blood loss, um how does that kind of translate to length of stay? Uh, what is the time and what's the cost savings to the hospital? Uh, what is the aggregate of treatment accuracy? So I think the good thing about um is that now we're seeing all this data slowly come out, and we're actually able to use this data to really prove the value of the things that that that that we do in this particular space. The third is that uh you could have the best um uh the best device, the best idea, but it has to fit into the clinical workflow within the hospital, and I think that's really important. So it has to integrate into kind of the um care imaging and hospital um operations into the clinical workflows. So with these three premises, um, a number of applications are possible in the hospital, ranging all the way from things like part replacement to device prototyping to medical education, dentistry, it's been well established. Um and what we are coming to see nowadays is uh a use of um 3D pending technology towards clinical applications for patient-specific use, such as in models, uh guides, appliances, and implants. So these are uh really um anatomic models and medical devices which actually are used for um the diagnosis and treat and treatment of patients. And and now we are seeing them in multiple different areas. So in our hospital, I think we've got um about 12 different areas and 40 different applications ranging from cardiac surgery to orthotics, really. And what they aim to do is maybe give a bit of a snapshot into some uh into some areas or where where we where it can be employed. Uh so one of them is a cardiac surgery. Um we we work in uh I work in an adult hospital, um, but we also see a bit of a spectrum all the way from pediatric to uh to to adults um in in the different hospitals within our group. Uh and we we've what we see is that uh cardio is really kind of changed the way that we we use to treat a condition called hypertrophic cardiomyopathy, which is a condition where there's actually enlargement of the ventricular septum uh causing uh heart failure in certain patients. And when um when when um it's progressed to a stage where, say, this thickening in the wall is is um it's uh refactory to medical management, uh what needs to be done is actually uh biomectomy. So the surgeon actually has to do open heart surgery to take off the excess muscle uh such that the patient's heart failure can be relieved. Uh so what we used what used to happen is that used to happen blind, so the surgeon used to just uh kind of take off as much as they thought was was adequate for the for the patient, um, looking at the pressure gradients at the end of the operation. Uh but what we're actually able to do now is that we're able to use our imaging and to define uh, say, the the normal myocardium, the one centimeter that we need to keep, and also the amount of tissue that we need to remove uh for the particular patient with modeling. And with this modeling, we can actually use this information and translate it in a number of ways. We can turn it into physical models that the surgeon can actually hold and feel, look in his hand and say, okay, I need to take out this much tissue. Um, and and that's really helpful for the patient. Uh, we can turn them into things like uh um rehearsal models. So for uh maybe a trainee who's done this the first time for a particular complex procedure, uh they're able to kind of localize the bulge, uh, they're going by a certain angle, say they're going trans uh transiortic, uh they know from a certain point of view how much they need to remove. And these are kind of done by silicon guides from the same substrate of using the scan and the modeling. Uh, we can also present information in different ways. So um now we've got things like uh um um different sorts of of uh of um uh surgical um um scaffolds that can actually uh curate tissue out. So in this is not so it's actually very important to kind of measure how much tissue you're taking out. So now we have we have ways to kind of uh weigh the tissue based on um uh the density of the tissue or look at displacement to see actually how much we need to take out. And I think all this really kind of translates uh the imaging into many different ways. Uh we use this for autopedic surgery as well. I think Tim talked about uh uh developmental hip dysplasia. Uh he talked about one method. So this is another method of the periacetablet osteotomy where uh you in this patient with the hips um uh are are slightly um um dysplastic, meaning to say that there's not enough um coverage over of the hip on top of the of the pelvis on top of the hip, and that might actually cause the hip to generate faster. So the aim here is to actually try to reposition the hip, uh, to offload the joint and to kind of increase the surface area on the joint such that the generation doesn't happen that quickly. It's quite difficult to do, uh, but with uh surgical planning, what we're able to do is we actually can define the cuts uh where this uh where where you need to kind of um um do the surgery to to to um to to isolate the acetabulum and then to um and you can do this by using the scan, um working with engineers, coming out with cloth with with uh cutting guides, and then using these cutting guides, putting them on the bone, uh anchoring them in place, uh, and by very accurate cuts uh that are enabled by basically the measured geometry that you can use from these jigs beforehand, referencing the surgery. Uh you're then able to uh make the cuts and you're able to actually put them in the take the different pieces and put them in the orientations uh that that you want. You can slide in and slide out the guides via K-wires basically. Uh and with this, you can actually uh really bring the engineering to another uh level of creativity by uh using the same guide wires to to kind of uh put cannulated screws within them so the same guide wire can be used uh for the screw. Um and and um and really this gives the biomedical engineer in the hospital uh or really with the company the capability to really put their creativity to how they do devices uh for uh for for really better outcomes. Um we use this for oncological surgery as well. So this is a patient with a sarcoma. Uh this is uh uh aggressive tumor of the uh sternum that's actually pressing the heart. This is a CT scan image, you see it all calcified uh here, and there's some activity uh in this on the pet meaning to say that there's metabolically active. Um surgery needs to take place uh to remove it and prevent uh heart failure and progression. Uh so what's um for these very complicated cases, uh you need really the surgeons, anaesthetists, oncologists, uh, cardiac surgeons, uh, orthopedic surgeons, resection surgeons coming together, and you need to get them to have really like a uh a global view of how the surgery is going to take place. Um, and really kind of having a model in hand with everybody on the same table allows you to plan this. You can plan the margins, you can plan uh how large the tumor is, you can have this life size. Uh what you can also do is that uh after taking it out, um it then allows you to work in different companies. So we work with uh a company called Ossiopol in Singapore. What they do is they produce uh beta TCP, PCL uh implants. Uh and what these implants do is that they can um we can then use them to uh correct the bone defect, um, to correct the bone defect. And really the surgical planning and the model helps really bring everything together. Uh you can actually bend plates on it, so you don't have to spend time bending them in terroc. Uh, and you you know that uh at the end of the day, the the operation that you do is accurate with the margins that that you want. Um and um yeah, this patient did quite did quite well. Um we use this for different applications as well, neurological surgery. So this is a patient uh with a cranial uh cranotomy, so that it had a piece of the skull taken out uh because of a stroke that happened. Uh sometimes you have to remove the skull in the initial phase of the stroke when there's brain swelling, but when the brain swelling is has recovered and the patient is better, you want to be able to cover the uh the skull again so the patient can continue ambulating and their daily activities. Uh so 3D printing allows you to do a few things with this. You can act and modeling, you can actually model uh what the defect looks like, how large it is, you can create moles. You can create moles in which you can actually put polymetal medal accolade, which is a common material used, a bone cement really, uh, that can actually fit um to the defect. Uh so we really the whole process of virtual surgical planning, using cats, so on and so forth, um, allows you to develop moles that can actually allow these to be uh molded exactly to the patient, a lot more accurate than than um the old than how they used to be hand molded last last time. And um I think what the hospitals are moving into um is really into peak implants. So the ability to actually manufacture implants at point of care uh is uh another frontier that the hospitals are slowly coming up uh up to. So this is not only um uh kind of like molding, um, but also to actually build the implants within the hospital uh itself. Yeah, so for all this to happen, I think um a number of things need to happen. You need to be able to kind of incorporate vertical surgical planning, medical device design, in-house production, and really integrate teams uh in the hospital within something called a 3D printing center or like a poll of call within the hospital for all this to happen. So I think um our hospitals department, and indeed many hospitals have departments where this takes place. Uh and for this to happen, you really need a confluence of things to come together. Um we we need to optimize the imaging for this to happen. So um for as as the the bills get more complex, um it's it's increasingly necessary to actually reach back to radiology to decrease uh say metal artifacts, decrease things that would make the the imaging less less optimal to reduce that, or say if you um you can say you can use your MRI. a different way to get your your your scans more accurate so you don't have to repeat a uh um uh uh uh uh another scan you can use the the um the accuracy of the MRI and and and and translate it to your surgical plan or there's sometimes um because uh by understanding the different modalities um you develop um uh a sense where you can um use the temporal resolution of your of your CT scan with your with your accuracy of soft tissue of your MRI and actually fuse them together uh for example that this is a patient with a kind of a metastatic uh codoma tumor in which a guide was was made for the patient so I I think it's really um being interdisciplinary working with uh specialties bringing radiology on board um allows you to kind of integrate different functions within the hospital then more so than one could do so alone um in addition the the a lot of engineering functions get into play so um your ability to use uh CAD to do spatial transformations Boolean operations is all very important for doing things like surgical guides um and um taking it one step further there's a feel of DFAM that needs to be kind of understood your ability to um to actually build uh around the product or and the technology you're using for so for photopolymers you will kind of need to put drainage channels in places where the anatomy is not there which needs you to understand where the anatomy is and where the drainage channels have to be or same for material ejecting uh that that comes with um uh materials that need to be removed so these have to be incorporated in certain ways away from the anatomy that you can kind of still see the anatomy but also um manage the engineering functions of it uh and with the engineering functions there's really a lot to know in terms of the mechanical properties of what you're building, compatibility, um dimensional stability to to come into what choice of material we're using and uh and uh and uh how do you actually design for this particular build that you're doing um and and all this is really kind of uh comes back to to really the ability for the the hospital to run the quality assurance uh and quality management system uh to to understand um really kind of the steps that happen during this process uh and to to mitigate um and to mitigate the um um areas where where where errors and and and lapses in quality food could happen and the important part of this is actually doing the design process so it's very important and uh to be able to kind of define the requirements uh review uh for uh review what the requirements are so this is a patient with a uh a bone spacer uh for a patient with a prosthetic joint infection they need a temporary uh spacer for the knee um that's uh that that could that that could be there for about maybe two months uh until the infection clears and the new knee could could be put in um and and with this um really with with with with with the bills we need to be able to uh verify the designs that come out in the end uh as you plan them and to validate that uh the designs work so we keep for example in this um in this knee space uh you know it's it's it's necessary to make sure that the material you use and and the way that you do the molding actually uh works at the end of the time and um it's necessary to actually kind of come together and bring the team together to do a rehearsal for for operations they've not done before so this these are some some surgeries that of um of periacetabular osteotomy that the surgeons will be new at we had them to come together do practices in the lab or say with hand surgery to come together and do practices in the lab for this to happen. And for this to happen I'm over to my last few slides now it really needs to be able to integrate kind of the knowledge of engineers and product design and process optimization and manufacturing with radiographers and imaging specialists and being able to um obtain optimize the image managers to kind of run the whole business behind it and also to optimize the processes and clinicians to to to provide a bit of the translation between all the different parties check segmentation and really um energize the whole ecosystem. And maybe this is my last slide I I think is also important for hospitals or people in the scene to really kind of reach back to all the different um stakeholders in

Funding And Reimbursement In Practice

SPEAKER_07 1:10:00

this fellow hospitals around the world academic institutions which do the uh which do research and RD in this area uh industry which which provides the tools for this to happen and regulatory which really provides environment to happen to really uh work together to to to to shape the conversations uh and to develop new capabilities for for patients yeah so this is my last my last slide uh and I'm happy to take um any questions uh mark um one question about reimbursement since you're in Singapore I mean the reimbursement is probably a national healthcare system um how do you get funding to create your lab and uh and and then you know going ongoing uh procedures yeah yeah sure so obviously for a journey um in brief it was funded by a by a grant so some of the initial work uh in terms of setting up the lab the lab initial hires was uh was was was was grant funded uh but over time uh what we have worked out with the hospital is a way to um charge uh for the work that's that is that that we do so so at the moment uh what we charge on is a function of the time that is uh spent uh um um by an engineer or a radiographer uh on this particular case basically uh so we we we convert this into a function of time and then we add uh kind of overheads and we also add um um material cost uh cost equation uh on a kind of a cost recovery model uh because what's um what what we are also putting forward is that uh the hospital uh by having such a uh service is able to do uh operations that they were not able to to do before basically or are they able to uh do certain operations faster so so um uh in in this we we look quite carefully at say what are the operations that actually have uh in the literature been helped uh by this particular procedure before deciding um what sort of operation to do or so on so on so forth so i i think uh these are some of the considerations that that that come that that come in in in into play uh with regard uh to this I could talk a bit more but basically that is the the the the the um the the the thinking behind this yes okay sounds good um all right we're we're tight on time so we're gonna move on to our next speaker mark if you can stick around that would be wonderful because I know it's kind of late for you um and we're gonna move around the globe actually Rashi even though she is located in the US right now uh Rashi Gupta she actually uh was in India so she has perspectives from both continents um Rashi I'll let you take it over she I'm not hearing you hear you're you're I don't hear she's not muted but I'm not hearing her um so hi can you hear me now?

SPEAKER_00 1:13:41

Yes perfect okay great yeah yeah so hi I'm Rashi I um I uh thank you Jenny for giving me this um opportunity and um so can you hear me yes okay I think it's a little um on and off um can you start?

SPEAKER_05 1:14:12

Um yeah okay just like I think your internet is spotty Hi um can you hear

India Vs US Adoption Realities

SPEAKER_05 1:14:35

me now?

SPEAKER_02 1:14:35

I can hear you.

SPEAKER_00 1:14:36

But when you when you start to share the sound kind of gone and I think your internet connection maybe okay um let's let's try it again let's try it again is it good now hold on one sec Okay you wanna just uh speak a few words okay good I think it's working now all right okay I'm sorry my zoom crash um sorry for the delay um hi Jenny thank you for giving me the opportunity to speak here um so I'm going to start with introducing myself and talking a little bit about what surgical uh planning is how it has evolved over the years and I'm also going to discuss a few bottlenecks that I have observed in my uh years of practice and where this um thing is going in the future okay so a little bit about myself I started my journey with medical 3D printing in 2019 um with a company called Imaginarium India Private Limited and um it is a 3D printing company uh and we used to cater to different hospitals and different doctors and uh predominantly we worked in orthopedics and we have worked with anatomical models surgical guides and um a lot of mock surgeries workshop and conferences and things like that. So I worked there for five years and I think I really got a nice experience of industry working in a 3D printing company. And since 2025 I have I am now based in St. Louis Missouri and um here I'm part of a 3D printing lab in a hospital Cardinal Jen and Children's Hospital and my job role is similar I'm still doing um 3D printing in medicine but uh the two the experiences that I've had are very different from each other. First experience was in India and it was in a company um so in this uh presentation I'm going to talk um a lot about the contrast in my experiences so whenever I mention the word industry what industry means is it is an entity which is outside the hospital it is a company like materialize uh which is outside the hospital and they are able to cater to different doctors and different hospitals uh whereas when I say point of care lab what I mean is a lab which a 3D printing lab which is situated inside the hospital now the advantage of um having a point of care lab inside the hospital is that um uh we are accessible to the doctors we can cater to different needs and we can really cut down on our um timelines right so yeah in um since I've started working uh in Cardinal Glenn Children's Hospital um one of the major differences between our practices in India and here is that a lot of my work here 50% of my work more than clinical work is about um researching about um providing evidences that this technology works about so we uh work with a lot of um different universities different researchers trying to prove about this technology rather than directly using it on clinical applications so um what is 3D surgical planning right uh 3D surgical planning is basically whenever there is a complex case uh a doctor can use these latest technologies um to plan the surgeries so that there are no um surprises in the OR right and so 3D surgical planning is moving us toward a world where every procedure is not just performed but it is precisely engineered and with help of 3D surgical planning you can actually predict what is going to happen in the OR and you can uh take uh um decisions based on that so how 3D surgical planning has evolved over the years okay so earlier um I mean um the first type of surgical planning is when you have a physical x-ray and the doctors are present in one room and they can just look at the x-ray and plan the surgery right they can use rulers to mark the angles and plan the surgery accordingly now in this case when you're planning a surgery on a physical x-ray um the output and the accuracy of the output just depends on of course it depends on many factors but majorly it depends on two factors which is how good is the image quality and um how good is the surgeon's judgment right so there are not many things that the output is is going to depend on there are only two factors. Now the next type of planning is digital planning wherein you have digital x-rays and you have software where um on the computer you can measure the angles and you can plan the surgery digitally. So in that case you are still planning on x-rays so uh apart from the factors that were already coming into picture a new variable a new factor adds upon which is if is the software accurate is the software calibrated right next type of planning is remote collaborative planning where um the doctors are not present physically in a single location so a doctor may be present somewhere else so you are using platforms like Zoom to um share the data and plan the surgery virtually together. Now apart from the variables that were there in the previous types there is another variable that gets added which is is the communication clear is the data sharing quality good are we sharing the data through encrypted proper software because all this data is patient specific encrypted data right so you cannot just share this data on WhatsApp or on any software. It needs to be properly encrypted before you share so as you can see as the type of planning evolves the factors that affect the final output also keep on increasing right um another type after that is 3D surgical planning which we do very often which is when you have a CT scan or an MRI you convert that into a 3D model or you virtually or you 3D print a model and then you plan on that model right so here apart from the factors that were there two new factors add up on that is the segmentation accurate is the 3D model that you have printed accurate so all these things affect the final output. Then you have AI assisted virtual surgical planning wherein um AI helps you uh you know plan the entire surgery and uh the doctor's inputs are uh minimal and uh and AI does most of the work and the doctor just has to validate. Now in this case um these software are of course made with help of a doctor but these software are made by data engineers or computer engineers. So in this case more variables get added which is is the software logic correct uh did the data engineer who made this platform does he know about medical has he made the uh you know proper um software so another variable gets added now the latest type of surgery which mostly we are talking about here is AR-based collaborative planning wherein um we have VR headsets and a doctor one doctor may be sitting in India another in US another in Belgium and they can wear those VR headsets and they can actually plan the surgery they can actually simulate the surgery using augmented reality right now in this case apart from all the variables that were there in previous um uh type of planning three more variables get added which is is the special alignment okay is the tech reliable is the system accurate um so as you can see as the complexity of our planning system increases the variables increase um but just because this AR-based planning has so many variables and this physical x-ray planning only has two variables is one thing better than the other is the main question right so even though um AR collaborative planning of course is very nice it is new it gives you more control um 95% of the cases in the hospitals are still done using physical x-rays right and less than 5% of the surgeries are planned using um ARVR and things like that. Now there are many issues why this type of advanced surgical planning is not uh able to scale so scaling is not a technical challenge right because we have the tech we have the software we have everything it is a systems challenge because not all hospitals are equipped with the correct systems to scale it and honestly not all surgeries need this type of complex planning right there are some surgeries which can be planned using normal X-ray in five minutes so you don't really need all these innovative and big systems for those planning. So it is really important to understand how to scale it when to scale it and is it even necessary to scale this thing right so innovation is fast. We as engineers can make 10 new apps in you know every day and I can innovate really fast. But the question is is the healthcare system going to adopt it you know that is the main thing. So as innovators when we think of a new thing um instead of just going on making new software new software new tech we have to also think about how useful is my uh tech how to integrate that in any healthcare system so in this presentation mostly I'm going to talk about these bottlenecks that come and why um adoption is a little slow. But before that um since I have had the privilege to work in these two different countries in these two economic different systems I would like to give a little bit of background about the basic things how these two systems work so that when I talk further you are able to understand my references in a better system. Now India is a cost-sensitive market right and uh still now of course we have medical insurance in India but for 39% of the spendings are done out of pocket. Whereas in US US is an insurance driven healthcare system most of the surgeries are uh covered by insurance and only 11% of the patients pay out of pocket. Now how how this affects 3D surgical planning is that since most of the surgeries in US are covered by insurance we need to convince the insurance people that you know this is uh this is regulated this is nice you should uh you know you uh you should get uh you should insurance and things like that but in India since the patients are paying out of pocket it is easier for us to convince the patient that this particular model is going to be helpful to you so you know you can buy this you can uh pay for this so because of that it is a little easier I would not say very easy but it it is easier to um uh bring the tech into the healthcare system because patient is going to mostly pay out of pocket as long as it is cheap as long as it is affordable for the patient um so I have also put some cost differences so like an MRI in India costs around 5000 rupees which is uh 60 dollars whereas an MRI in US costs 1500 to 3000 dollars right but this is not a very huge number because for a for an American for a person who has health health insurance this will be covered by the insurance so they'll have to pay zero right but in India this might not be covered by insurance so we'll have to pay it out of pocket but it is not a very huge amount that we can't pay. So now if I have a 3D model which I am going to sell in $10 what is $10 uh you know with a $60 MRI so it is easier for the uh people to use this and to get for us to get paid for it whereas in in US we have to really um you know have to have uh uh proper pathways proper insurance pathways for it to get reimbursed okay um so now I'll talk about some of the bottlenecks that I have personally experienced in my um seven years of medical 3D printing um the first and the major bottleneck is the communication between a doctor and an engineer so this is this is life case this has happened with me in 2019 when I started doing this um so uh an orthopedic doctor who I used to work with Dr. Taral Nagta he came and told me that uh we have to do a general surgery do from the distal end and just align the mechanical axis now for me as a biomedical engineer who has only um learnt about computers and maths uh didn't know all these values so all these terms so I had to really work hard I had to read the orthopedic books understand it and then start translating in my mind what all those things mean. Now when um I have had juniors when I have had to train someone now I know the language of engineers like you know how to translate it. So now when I'm trying to extend to an engineer I will say that you have to make the bones collinear and remove a triangle and the axis should be collinear or parallel right so this language now the language that I've stated here doctors engineers don't know this language and doctors don't know this language. So the biggest barrier the first barrier that we face when Working in this industry is this communication, right? And this communication gap, of course, of uh results and delays, and then um point of care, especially like in a hospital. We really in a hospital we really struggle to hire a person who will know all these things. I think it is easier when you work in an industry because um they have people who can train, but in the hospital now, since I'm working in a hospital, I don't have any senior of me who knows the engineering language and who can help me translate, right? So, solution is to have um residency or fellowship programs that uh teaches all these things because earlier there was no stream that actually joined these two industries, but now there is. There is a lot of research, so we really need um clinical engineers instead of biomedical engineers who properly understand both these languages and can translate between the two. Another thing is awareness. Now, since we are situated inside the hospital, um it is very obvious to think that of course everyone in the hospital knows that we have a 3D printing lab, but that is not the case. We we meet many nurses, many physicians who don't even know we exist or who get confused with uh from uh uh 3D recon lab. And they're like there's a lot of uh issues because we have to talk to people, we have to explain them what we do, we have to drive them in and things like that. And also when it comes to industry, like when I you when you are working in a different company, since you have a marketing budget, since you have marketing people, salespeople, they will come and talk to the surgeon, and it is easier for them to infiltrate than us, uh, even though we are situated in the same hospital because we are like one or two person team. We have to do the marketing, we have to do the sales, and like everything is only us. Right. Um, so of course the solution is uh solution to this problem is we have to do internal marketing, we have to uh put up case studies, um, platforms like 3D Heels or materialized forums where we talk about these things so that doctors are aware that uh, you know, uh point of care labs are there and and it it will really help. Another major issue, this is a very funny issue, is um can we get paid only for planning, right? Um so this is one of the cases that I had done in India where the doctor didn't need any surgical guide or anything. Um he just said that it would be great if you can plan the surgery so that I can uh you know follow that surgery in the operation theater. So I planned the surgery, I took big printouts and I stuck on the uh OR uh wall. Now, this may look like a very normal thing, but trust me, this is very helpful for the doctor because these screenshots that I have taken is in such a way that the doctor uh can actually see this type of orientation in the CM inside the OR, right? So when he's operating, he can directly look at these screenshots and understand how I want to put the wires. As you can see here, I have tried to make the bone transparent and I have tried to see from where the uh K wire is going and everything, how the implant should look. Um, now when the surgery was over, I collected all these things and I went to the insurance office that I had done the planning for the patient. Can we get paid for it? Like, can my company get paid for it? Um, so there was a resistance because they were like, this is just a piece of paper, how can we pay for this? Um, so you know, uh when you have something tangible, when we have a 3D printed model, we take this and we say we are charging for this model. People understand, okay, this is the model, okay, we will uh pay for this or something like that. But what is planning? Like uh we should get paid for it, right? Um another example, this has happened here in the US when we were working in the when we are working in a hospital, that you are working, you meet a doctor in the hallway and they say that um, okay, can you quickly plan this surgery? You just have to take this, this is and they say it's it took five minutes of their time to explain the surgery to us. We took five minutes to plan the surgery and send him the uh screenshots and he used it in the surgery, everything happened. Now we took five minutes because we have had years of training, we know how these things work. But should we not get paid for those five minutes, right? So that is why many insurance companies we have had this resistance that they ask it is only planning, and doctor said what to do and you did it. Um, should you even get paid for it, right? Another thing is that of course we are salaried, we get our salaries, but the hospital is the one who is um absorbing all these costs, right? Hospital has to get paid for the planning services that we are providing. Um, so this is also one of the major issues in surgical planning because we don't have anything tangible that can be charged for. So there are solutions because there are new CPT codes that are coming. There are surgeons who uh while filing for the insurance have to mention very properly that I use their planning services, so they should get paid for it. But this is one of the resistances that we have seen.

Speed Versus Standards In Patient Care

SPEAKER_00 1:33:06

Another and major um resistance is speed versus standardization. Okay. So to put across this point, um, I'll talk about two scenarios. Okay. First scenario is that a surgeon comes and he says that uh the surgery is tomorrow. This is a CT scan. Can you plan the surgery for me? Or can you make a 3D model? Now, of course, we can make, right? But if he comes with a um rose low resolution CT scan, I can do the planning through it. I can do everything in less than 24 hours. But the question is, should we? Should we do it? Because if um, and I am sure whatever planning we do with the low resolution um CT data, it is going to be helpful for the for the surgeon. He's going to um use it properly and the surgical outcome is going to come better. But the question is, should we do it if the CT is low resolution? Because if something happens, then who is accountable for it? You know, I chose speed instead of standardization, and if something happens, then what happens next, right? Another scenario is a surgeon comes and says, okay, just make a 3D model. Now this surgeon has not told us what is the application of the 3D model, what is the timeline, what does he want, when does he want? So now if he's asking for a model just to show it to a patient, right, then I can just uh put a junior uh to it, he can make a model and uh the I can print it in a cheaper material and give it away. But if that model is going to be used for a surgical planning, or if it is uh, you know, model is going to use for plate bending or something more complex, then I will have to have more time. I will have to put a senior engineer to do this, I'll have to do more steps of validation. So, how how will you decide should I uh choose time, should I choose standardization, should I choose resources? And that is the reason where uh why we should always go for standardization, because uh, you know, we choose speed sometimes, we try to do it fast, but then um things can take a wrong turn, right? And that is why the cost and speed versus validation standard is an ever going um argument and validation should always win. So these are some of the companies who have FDA approved software that um have all this validation and standardization done. Now, of course, these software have slower workflows because they are situated outside the industry. So uh doctors have to go there, um, like doctors have to talk to them, get the services done. It is a little difficult uh expensive, it is a recurring cost, but that will make sure that each output that you get is trustworthy, it is repeatable, it is safe, and it is regulated. Sometimes you have to choose speed, you have to choose uh cost when um it is only for patient education or if it is only for uh, you know, doctors like to keep some models on their desk so that they can um talk to the patients about these things. So at that time you can choose speed, but um, as a thumb rule, it is very important to choose validated proper software so that you don't um face any issues in the future. Um so those were some of the bottlenecks and um about future direction where I think this industry is going. First of all, um I have talked about some of the solutions of these bottlenecks in the uh uh slides itself, but a hybrid workflow is very important. Uh automated translation, right? Like I showed in the slide, the uh language between the surgeon and the engineers. If we have something like a Google Translate that can help translate these things, which is now very simple using AI, uh that will really help uh bridge this gap. And if there is AI assisted segmentation and planning, wherein uh which is available, materialize has it, many uh uh many companies have it. So that those things are really decreasing the time um and increasing the accuracy of these systems. They have to be standardized protocols, and uh the integration in the hospital system has to be high. And of course, the future uh thing is uh uh if we have a digital twin of the patient so that we get the feedback uh of our model and then we can work on it on getting better. And uh yeah, another issue is that industry is move moving much faster than evidence, right? So, for example, um back in India we started uh doing a study on Club Foot. We made 3D printed Club Foot uh braces for uh uh children in 2019. Now we were waiting for two years to get the output, to get the result, to see how our braces are doing. But till uh 2021, we were still waiting for our validation results. And if we see in the industry, there are hundreds of new club food braces that have come. Whereas we were still stuck to our 2019 design because we were waiting for clinical validation of our previous base. So these things actually demotivate uh the industry to try new things, but uh that should not be the case. I understand that industry is moving faster. There are innovators who are making new um software each day, but we have to stick to what we are uh doing, we have to wait long term for the outcome results and make sure that whatever we are using is validated, is standardized, and repeatable. And yeah, the next phase of 3D surgical planning will not be designed uh defined by better technology, but by better systems. Um so that's my time. Thank you so much.

SPEAKER_02 1:38:57

Excellent presentation. Actually, um the audience who really loved your presentation is uh very real. Um let's see, do we have any questions? Um okay, one question from Caesar in your experience. How has it been in terms of the journey mixing 3D surgical and materialized software? So I assuming you use a bunch of different softwares for your workflow.

SPEAKER_00 1:39:19

Yeah, but yes, yes, yes, definitely. We have uh used uh 3D surgical, we have used materialized software, and um we were one of the first people in India to have materialized mimic software. That that has been amazing. Um 3D surgical segmentation software also is uh we we um back in India, I think in 2020, we uh 21, we also helped 3D Surgical to um when they were launching their software uh to uh help and validate the software. So yeah, our experience has been great with both the software.

SPEAKER_02 1:39:53

Okay, fantastic. Like I said, we have probably a lot more questions for you, but we are running on out of time as well, and we have one final uh speaker. So uh let's uh welcome Dr.

AR/VR And The Digital Twin Path

SPEAKER_02 1:40:04

David Purston.

SPEAKER_04 1:40:06

Hi, thanks, Diddy. I as you saw there for a second, I apologize for interrupting with my video for some reason upside down.

SPEAKER_02 1:40:13

I know. Well, you know, I think people are paying more attention now, you're upside down, but uh hopefully hopefully your presentation can still show how so let me uh go ahead and do this, and uh hopefully that will be uh much more interesting to look at.

SPEAKER_03 1:40:28

Uh and I think I have this right.

SPEAKER_02 1:40:32

If we do this and okay, so Alejandro, you're asking about where David is. He is in Connecticut, US, not Australia.

SPEAKER_05 1:40:44

Are you in Australia?

SPEAKER_03 1:40:46

Me?

SPEAKER_05 1:40:46

Yeah.

SPEAKER_03 1:40:47

No, I am not in Australia.

SPEAKER_01 1:40:49

No, no.

SPEAKER_04 1:40:50

I apologize. Sorry, I know I just turned it off. Let me do that again. No, I am in the United States.

SPEAKER_02 1:40:56

Okay, sorry, it was a joke. Sorry, it was a joke. No. Okay, can you can you share? Because we only have like less than 20 minutes now.

SPEAKER_04 1:41:05

I guess uh working on it. Uh I apologize. Um let's Okay, something's working. How is that? Can you do you see my screen?

SPEAKER_01 1:41:13

Yeah, I do.

SPEAKER_04 1:41:16

Okay. Okay. And you can hear me. Okay. Great. So okay. We'll try and get this done. I won't read quick. I have um uh a lot less stuff, so we'll move quickly. So thank you very much. This is really exciting. I have to say, I really enjoyed all the presentations, and honestly, it got it's almost like a perfect setup, a really great segue. I mean, you ended saying the words human digital twins, and that's what I want to talk about uh to a large degree. And I I I just want to thank um and brink a couple of things that came up because they really resonate in this. I think Jenny you like this. It's all gonna come together, all this previous stuff. Beatrice had talked about volume renderings versus you know uh segmented models. I I I love that, I so completely agree, and I I love that you're you know that are focusing on it and and the value of a 3D model. So I'm a surgeon uh by by training and by trade, and um I I totally get that. And then Mark had talked about workflow, and and I something he said is like a literally a quote from meetings we've had where if you can't fit into the workflow, it doesn't matter how good your product is, you've got to know you know your customer's workflow. And then he also talked about optimizing the imaging for 3D. I love that too. I I saw a recent piece from C and it's that their hardware is is now really realizing the fact that you know these things are being done in 3D, so how can we optimize the image intake? And a lot of the work we do is optimizing the file for 3D representation, particularly in in ARBR visualization, which we're gonna talk about. Uh and then finally, Rasheed, um uh you know, talked about uh uh a few interesting things. You showed a picture of 3D virtual surgical planning, a couple of surges, and it was an AI image. I'll send you an actual real photo of some of our uh users actually doing it for real. I mean, this this is actually going on uh for real all the time. Um I loved your slide about awareness and people not knowing that there's a 3D lab right down the hall. My God, we suffer from that all the time. Um it's um and the idea that let's let's really branch this out and call it a 3D lab because there's so much more to this than just necessarily making a print. That's a huge part of it. But there is so much more to this. Again, we're we're we're all uh agreeing the same thing. Um and and that was my other big conclusion. I'm sort of preaching to the choir. I get that. Um, but you know, that's okay. Um all right, so let's uh let's talk about uh the uh human uh digital twin. Um just to again talk about something that everyone's been saying here. If we look at this from the center, let's talk about the virtual model, right? And we all know how we get there. We start with the CT, the MRI, and then we do segmentation. Often there's a lot of AI involved there, a tremendous amount of fascinating things going on. Um we get to this virtual model in the middle, and then what do we do with that? Well, we can print it, we can visualize it, we can share it, we can do all sorts of very interesting things. So we live over here in the visualization part, but more and more we are really focused on in talking about the concept of this virtual model and that it is a human digital twin. So um, here's my my sort of my spiel about that from the surgical standpoint, which is that we don't do this in in medicine and in surgery, and it's it's it is a little bit weird. Um, we're really the only industry um that is uh you know a high-stakes, uh, high technology industry that does not use simulation in any significant way. You know, how many surgeries go on every day all around the world, and there's really not some any sort of practicing or simulation of the surgery. What we call pre-surgical planning um is is really that in a very different way. And and it's really sort of a structural gap in how all of this works. Um, so how uh how can we approach this? And and so a little bit of background, you know, we've seen this progression in most other high-tech industries. So Honeywell was just sort of a very good example. They started out, um they were actually building physical systems, right? They built hardware controls, physical buildings, infrastructures, things like that. Then they sort of transitioned into the IoT part of it, the platforms that were connecting things. And now what they're doing is they're focusing on actually just building the digital twins of those things, and then just working on that as a platform. So, this whole idea of like the platform is what we're talking about, and we're gonna put human digital twins onto a platform. We're all on this same journey. So um it's it's interesting. Um so specifically, our come our uh company, DICOM Director, um, again, right now, what we do is take the the models that we're talking about and we bring it into ARBR, and all the interesting things you can do with that in terms of it being a medical simulation. But it where we want to talk about is in the future, where this can go, and and the the first step, and we're already looking at this, and I know other people are as well, is starting bringing this into merging the the two worlds, um, the real world and the virtual world in in uh training platforms, and then ultimately getting to where there's actually a consumer market even and and telehealth and the idea of the human digital twin and what people do with their own human digital twin and things like that. It starts to get into some very interesting areas. So here's just a quick view. I apologize, I don't I don't have one loaded up, but I would ask anyone who has an interest, go to our YouTube page. There's a lot of videos there uh of a lot of our use cases. This is a case that we did recently, it's a Tetrology of Philot uh case, um, which was was pretty fascinating. Um and again, the full video is uh is up there on our YouTube page. Um this is uh pointing out some of the defects. Um this is all filmed through an Apple Vision Pro. Uh uh we were talking previously about some of the hardware. Uh there are um there's a lot going on in the hardware market right now for AR, VR headsets. Uh all of it is good news. It's all happening very fast. There's a lot of changes uh going on, uh, but it's all gonna end up with us uh having uh better opportunities in the future. Um let me move on. With these technologies now, we're able to uh so this um oh we're gonna sit through the whole video, uh just a good example of um the patient experience uh about this. Uh it's really sort of the low-hanging fruit. We have a lot of other interesting applications, but just the idea of putting a headset onto a patient and allowing them to understand what's going on in a way that you never really have before. And what's what's nice about the technology is we don't have to go um all the way at the beginning. You can start with is something as simple as a a patient can wear a headset and just be looking at the anatomy by themselves. You can have a projection system where you've got a doctor or somebody, a nurse, an assistant, wearing a headset and projecting screen sharing it to a large screen, and uh the number of people can be watching in that, or the ultimate, as some other people had referenced, if you get into a virtual collaboration where you can have multiple people all in the same virtual environment. And those people don't have to be in the same room, they can be anywhere in the world. Uh, they appear as avatars to each other, and all can see the model and all can control the model. Uh so you end up with uh a really a lot of fascinating ways for uh not just the pre-surgical planning um but the the patient experience and patient understanding. We we love this particularly in the pediatric cases because as a surgeon uh I found there's a lot of variation in what people want to know and how much they want to know and understand in things in pediatrics. The parents want to know everything, they want to know every detail as much as they can understand. And this has really been a big a big breakthrough uh we've seen with a lot of our customers.

SPEAKER_03 1:49:24

I'm sorry, let's see. Next slide. How do I get to the next slide? Oh, there we go.

SPEAKER_04 1:49:30

Um quick thought about hardware. Again, happy to talk to anybody who has questions about hardware. There is a lot of hardware out there now in the ARVR world. I'm very happy to say we run on everything, so uh whichever one you decide you want to go with, uh, we uh can can make that work for you. There's advantages and disadvantages to different things. Um it's a bit of a long conversation. Um maybe one day, Jenny, we can have a whole session just on that. Um, but uh please feel free if you've got any questions about. We have a lot of experience. We have demos available on everything and headsets that we can demo with.

SPEAKER_03 1:50:06

So that's that's a another interesting component about this. Whoops.

SPEAKER_04 1:50:13

Um getting into some of the interesting use cases and and some of the AI uh um and where this can go and the idea of the human digital twin and what it really means when we talk about human digital twin versus just oh, here's a cool 3D picture of a CT scan, is when we looked at predictive analysis. Um and there's a lot of interesting things you can do in my area is oncology. So both the idea of a uh tumor mass expanding and what is going to happen if that happens, and then also preferably what happens is seeing a tumor mass shrink over time and what that can mean. Um this is provocate not only in terms of the patient understanding, but uh from a physician and a care planning standpoint, I think this is really gonna be quite revolutionary. And I think you're gonna see that in treatment planning meetings, you're gonna see 3D presentations of, well, this is what's gonna happen if this continues to grow in this direction, and this is what we're gonna need to do in this period of time. So it really starts to change the role of our imaging now, instead of just sort of here's what we have now. We can start talking about, and here's what's gonna happen in the future. And and again, human digital twin, this is what you can do with projections. This is an interesting way uh to to look at this. It's a it's a bit of a I think this will speak to the biomedical engineers uh in the group, um, using the term a stack, a technology stack. I'm a doctor. This uh this is something I had to learn. But um, but if you look at it this way, it's really sort of interesting because all of us here in this business, we we fit really into a really important part in this, and outlined here in blue, which is what we would call the spatial computing uh part of this. It all starts at the top with the CT scan. And again, we said there's a lot of interesting things happening there from Siemens and GE and Phillips and what they're actually doing with the image acquisition and how that's being optimized for uh three-dimensional uh modeling. And then a lot being done with how the data infrastructure is being used, and then of course the AI analysis and how we're doing the post-processing. And then it comes to us, all of us, where we do the segmentation and we bring it into AI or we print it or whatever, and then we give it over to the doctors and the planners and the people like myself who are out there doing it, and then of course there's the hardware that plays in. So it's really all part of a whole uh ecosystem that I think is the future of how imaging is used in in healthcare. And I think we are up to yep, that's it. That's um uh kind of the whole thing. Um, and if anybody has any questions, let me see if I can't make my awesome.

SPEAKER_02 1:53:00

Uh we're like cutting really close, but we made it on time. Thank you so

2026 Goals And Closing Notes

SPEAKER_02 1:53:05

much for a great presentation, David. Um, I think we should do a podcast uh at some point to talk about this even more in depth. Um I would like to invite uh actually everybody on this panel to do our podcast. We can have a longer conversation than just this is really what a preview of a very complicated ecosystem. Um and also I also want to direct people um to our YouTube channel and Instagram and podcast, because we have multimedia way of presenting the same set of information today. But thank you so much for uh for for sit with us so far. So we have a couple minutes left because I have a hard stop in five minutes. Uh let's have all the panelists just to be um back on. I'm really appreciative that you guys stay till the end. I hope this is as useful to you as for the audience. Um since we don't have a whole lot of time, do you first of all, do you have any questions for each other? Okay, great, David. Um I think you got my attention. Um there you go. Okay. Um so yeah, so do you guys have any questions for each other? Uh I want to give you an opportunity for that. Oh, you're all muted for some reason.

SPEAKER_04 1:54:17

I I'm planning on reaching out to virtually every single one of you.

SPEAKER_05 1:54:21

Awesome.

SPEAKER_04 1:54:22

So because I I every I have notes from everybody about what you're doing, and and it's just so wonderful. And I really love encouraging this idea of this ecosystem because um yeah.

SPEAKER_02 1:54:34

Okay, and um some people are asking for contact info. Uh, first of all, I I would say it's helpful for people to actually share who they are asking for contact info. So that would give you some people some context. So please do so in the chat if you want to share your contact directly to the audience. Feel free to do so, social media, whatever. Um now, okay. I have one last question is what is your immediate goal for the rest of 2026? Who's ready for for this?

SPEAKER_04 1:55:03

Oh, I got an easy answer for that one.

SPEAKER_01 1:55:05

Okay.

SPEAKER_04 1:55:06

We're fundraising.

SPEAKER_01 1:55:08

Okay. We should talk about that. Okay, Rashi.

SPEAKER_00 1:55:12

Um, my immediate goal right now is uh reaching out to more universities and more students, talk about this thing so that uh, you know, while they are studying studying this, while they are in academic academics, they can learn more about this and be more equipped when they come into the industry. They come with uh because for me it was a bit of a learning curve trying to understand. And I really wish while I was doing my bi biomedical engineering, I would have had um someone to tell me, please read these orthopedic books also. It'll be helpful for you in the future.

SPEAKER_02 1:55:43

So yeah, let's definitely continue that conversation with 3D Hills. We aim to educate, so uh we have some ideas, and I'm looking forward to your blogs for us as well. So everybody can stay tuned for that. Uh and Beatrice, do you have anything for the rest of 2026?

SPEAKER_08 1:56:00

Yeah, the the the big focus this year is for sure on collaborating with reimbursement. There uh I mentioned it here the we are supporting, of course, uh the registry that is going to start soon, but also in Europe there are several initiatives. And uh I think that uh we are a bit privileged in that way you're having this global perspective where we can actually uh share the knowledge from some countries to others and trying to bridge the gap so that we don't need to reinvent the wheel everywhere. Um of course, every system is different, every system is going to work uh in a different way, but uh the the clinical evidence it's it has to be there for everyone. There are commonalities, so learning from it, trying to help um the community to build that evidence to get into reimbursement is for sure the goal.

SPEAKER_02 1:56:49

Yeah, I have to say I'm a little out of date on the reimbursement side and actually working on a podcast focusing on reimbursement for MetTech. So stay tuned for that. Um okay, anyone else? Tim, Mark, what's your 2026 goal?

SPEAKER_06 1:57:02

Oh, so we have two major milestones. One is uh 510k clearance for the software uh that we want to achieve, and then the other one is uh uh to complete the first cohort of five patients in the clinical trial. Uh tweet the first five patients and get six months uh follow-up to get those first uh very important clinical results.

SPEAKER_02 1:57:22

Yeah, love to hear more when's those data ready.

SPEAKER_06 1:57:26

Okay, sure.

SPEAKER_02 1:57:27

Mark?

SPEAKER_07 1:57:28

Yeah, first of all side, uh maybe three things. One of them is to increase outreach. So we're looking to uh basically go do more cases, uh not only in a hospital, but in the enabling hospitals. Uh the second is to have more do more applications, try to um expand our applications. And the third is that uh we are looking to implement a quality management system. So we are we're really kind of banking on on on on um it's a long process, but we're we're getting we've done a lot last year and we're going moving more and more uh forward with with each year with the quality management. Yeah.

SPEAKER_02 1:57:59

I got an idea. We should do a podcast soon because I know I know you've got approval. So we're gonna do a podcast soon and then you can share that podcast to your hospital. Okay. Good. Okay, thank you everyone for joining us. We did run out of time a little bit, unfortunately. Uh I hope everyone has something to take home with. And I will see you soon. This webinar will be repurposed on our YouTube channel. We will reconvert it into podcasts so you can listen and reach out if you have any questions or comments. I'd love to hear your feedback. Okay. Well, thank you so much. I'll see you another time.

SPEAKER_03 1:58:34

Take care. Thank you so much.

SPEAKER_02 1:58:35

Thanks, Jenny.

SPEAKER_05 1:58:36

Thank you. Bye bye. Bye.