So I'm Russ Miller, and along with my, with Chris Manley, we're the, he's the chair, and I'm the co-chair for this, the ABIP sessions. Today we're going to be talking about, about imaging in the bronchoscopy suite, and how to get the most out of it. Our first speaker is Joe Sicinia, who is a advanced diagnostic bronchoscopist at the Cleveland Clinic, and one of the people that's really advanced the field, and in our knowledge and understanding of how to best utilize imaging in our bronchoscopy suite over the last few years. And he will be discussing how we can engage the full potential of our standard bronchoscopy systems. Hey, thanks, thanks Russ. I can't pretend to teach you guys how to unleash, how to unleash your fluoroscopy in 12 minutes, but I'll do my best. But if you want to learn how to unleash your fluoroscopy, I invite you to Cleveland. You could spend some time with us, and you could see what we do. There we go. Okay. So, thanks for the introduction. So, just some disclosures, really for this talk, just two, body vision, no medical, and you could see what I do for them. Mostly, mostly product development. So, this is obviously, you know, this is the nice nodule. This is what we're all hoping for, you know, in the bronchoscopy suite. But unfortunately, even images like this have their pitfalls, right? Because it may not be what they seem. So, the, you know, the, what we get is not so much the guy and the dog, it's this, right? So, you think you're on it, but when you rotate out of plane, you could see that you're clearly, you know, needle's not in your lesion, right? It's off of your lesion. And I think as bronchoscopists, and as we train, we train in the AP view, right? We train in coronal views, and we don't really train and learn how to interpret imaging and live fluoroscopy in oblique views. And the oblique views here are obviously very, very important, as we'll talk about, you know, in a second. So, typically, you know, in the bronch suite, you have your patient there. There's your nodule. The arrow is your needle. And your view, basically, is AP, right? So, there's your C arm there, and we, for our, you know, for most of my career, you know, growing up, you know, you're in the AP view, but that could be misleading, because that, although that AP view, when it projects, it looks like, you know, needle and nodule, right, to the right. That's the representation of the nodule with the needle and what you're seeing on the fluoro screen. When you actually move your fluoro into an oblique view, you could see how the needle then just spins away, okay? So, a good technique to learn is when you're doing your biopsies, when you're doing your navigations or whatever, move your C arm out of AP, move it into orthogonal views, okay? RAO, LAO, so on and so forth. You could put a little thing together here. So, if your nodule, say, is in the right upper lobe, and you turn your fluoro obliquely into RAO position, and it does not have to be a big RAO, 10, 15 degrees, and the needle in relation to the nodule moves right on your screen, then you're anterior to it. If it moves to the other side, you're posterior to it, and the same for things on the left. Now, what do you do with that information? That's a whole other thing that the other guys are going to be talking about with robotics and image targeting and stuff like that. So, just a little primer about x-ray-based imaging, right? So, we all know what conventional CTs are. They use what's called a fan beam, right? There's a fan beam emitter. It's spun around the patient in a helical pattern, right, with constant detection, right? And it creates a nice 3D image, but what fluoroscopy does is uses a cone beam emitter, not a fan beam, a cone beam emitter, which creates basically a 2D image, and then, if you could take that cone beam emitter, spin it around a patient, say, to around 200 degrees, it could create a 3D image. I think A.B. is going to talk a little bit more about this later, but this is the basis of what's called cone beam CT. It's basically a cone beam emitter, 200, you know, a wide angle of scanning to create a 3D image. Digital tomosynthesis is the same thing. It's a cone beam emitter, but instead of going for wide angles, it goes over shorter angles. So, somewhere around, say, 60 to 80 degrees, okay? It's the same process. You have an x-ray tube that moves around the patient, but a smaller angle, less data, less data to reconstruct, okay? You'll get decent images around the isocenter of what you're spinning around, okay? So, if the nodule itself is in the center of that spin, you'll get nice images, but everything behind it will be kind of dull, and you could see a representative cone beam image above and a representative digital tomo image below, and the, you know, kind of the photographic analogy of that is what we see here. A CAT scan will be on the left, where you could see everything in the foreground and background very, very clearly, okay? Because you have this wide sweep, lots of data to reconstruct. It becomes, you know, it becomes a true image. But on the right side, you could see that only the dog in the foreground is sharp. That is the isocenter, and everything behind it, because you're missing data, because you're, you know, missing things to reconstruct, it becomes very blurred. So, the analogy to us, in our suite, right, is to isocenter the nodules. Make sure the nodule is in isocenter, so when you're doing your spinner, you could see your nodule really well on the reconstructions, the 3D reconstructions that you do, because if you're outside of it, your image could be blurry, if not seen at all. Many technologies that use digital TOMO will provide workflows to approximate isocenter for yourself. Some use, some use main carina to identify isocenter. Some use the scope, so you'll navigate out, and the scope will get close to the nodule, and then you know that you're sort of approximated to isocenter, and then you kind of do your spins there. The other thing that you should be aware of in your Bronx suite, is that there are two types of fluoroscopies. There's the flat panel, which is a digital detector on the bottom, you know, the Maserati of detectors, and then there's the image intensifiers, which was, you know, the old historic type of C-arms. The image intensifiers generate analog, an analog signal that then they convert to digital, and you get an image, but not to get too much into the physics, the flat panel is a much crisper and clearer image than you'll get with the older image intensifiers. There's workarounds to, you know, make the image intensifier a useful image, but it takes algorithms to do that. The other issues with image intensifiers is that it gives you this pin cushion distortion at the outer edges of the image. So the center of the image is nice and crisp, and as you move out, you sort of get this distortion, and the flat panels are undistorted. The other differences between the two is that the image intensifiers will have a smaller field of view, so it's a rounder, smaller field of view versus the flat panels, which are larger field of views, and it requires more energy, more radiation to give you the same image with the image intensifiers than it does with flat panels. So the factors that will affect your DT imaging, as we talked about, are the detector, right? So a flat panel versus an image intensifier. The algorithm that you use, and as you'll see when you look at technologies that are out there on the floor downstairs, you'll see some of them have really great algorithms that reconstruct things really well, and other ones that kind of do a decent job. So the reconstruction algorithms that are inherent to the technology that you're using are super, super important. And then the sweep angle. The larger the sweep angle, the more data you have, the better the images you could reconstruct. I just want to touch on this. Roberto's going to talk about this later, but radiation exposure. Again, cone beam CTs, lots of radiation. Digital TOMO, much smaller radiation. You could see it's a factor of almost 10 on the left panel. So digital TOMO, for our purposes, we don't need wide angles. We don't need to look at the whole lungs, like, you know, in cone beam CT. For what we do, we just need that local area of where the nodule is. And if we do it using DT, we could get by with a lot less radiation. So what else could digital TOMO do? Well, it certainly identified where the nodule is, but there's other things it could do, like, you know, using the, you know, whatever algorithm you have in your technology, you could use that information to correct for CT body divergence, which is a whole other talk in and of itself, we all know what happens. But that they could use the information from DT to get there, okay? So, and one of the things that they could do is take advantage of is augmented fluoroscopy. Now, the lights are on, you can't really see it well, but the top image is a nodule, and you could see the needle coming out, and it's deflecting underneath the nodule. Now, here you could see the nodule, but just imagine in your head that this is, you couldn't see this nodule at all. What we could do with DT and the algorithms that go with the technologies here is to put an augmented fluoroscopic overlay, right? So, a fluoroscopic overlay to show you exactly where the nodule is. So, when your needle comes out, you know you're getting it into the region, into the region of where the nodule is. Again, you have to imagine here, here obviously you could see it, but imagine you can't. So, this information could help you, you know, target your lesion better than you could have before, because you could have even gotten into the right spot, but your needle gets deflected at the last minute, you get a non-diagnostic result. What you just saw was the NOAA product, this is the body vision product. Again, very similar, you have an augmented image, an overlay, and you could spin these overlays in any different direction. You could see it in the top panels, obviously the coronal or the AP, but you could spin into LAO, RAO in all different directions. And again, you could take advantage, especially with the catheter-based technologies, you could take advantage of what we talked about before. If you do a spin and the catheter separates in any way from your nodule, you could approximate if you're in front or behind the nodule as well. The other thing that you could do with this technology is tool and lesion, right, which we all know is now like super important, it's like the trendy thing, right? It's like Taylor Swift. Sorry, I can't believe I mentioned Taylor Swift in the talk, I'm so embarrassed. But, you know, look, tool and lesion is super important, right, that's what IR guys do, right, for TTNAs, you know, they're successful because they ensure that their needle is in the nodule and that's what we're trying to do here. So you could achieve tool and lesion with digital tomo very easily. And you could see representative images here and here of how you could do this. These are all DT images. This is the NOAA product, again, this is a shorter angle sweep, but again, this just highlights that you don't need a large angle with big detectors. This is a small, this is a short angle sweep, a 60-degree sweep with an image-intensive IRGE9900. And you could see there's needle and nodule on the left and you could see the representative CT image on the right. Is it perfect? No. Is it adequate? Absolutely. There's also other things that you could do with images using algorithms and AI. And what LungVision does is they take the DT images, which you see on the left, and they could apply other filters which will remove some of the background vasculature. They'll accentuate the nodule itself and they'll accentuate where your scope and needle are. And it's just, it's all algorithm-based. It's photoshopping basically the image. And I don't mean that in a derogatory way. It's just, they're just pulling out what's important to you. And you could see a difference between the images on the left and the images on the right, which is what they call AI tomography. And, I mean, the images that they get are really great. This is with the SIOS spin, which is probably the best, right, the best in, you know, one of the, you know, the Maserati of what we use. And you could see that's the preop CT on the left and that's the product that you get on the right, you know, after the algorithm and your spin. And here's your GE9900. I mean, this is an analog, you know, image intensifier. And this is the output that you get. It looks pretty similar. So, and then this is just the GEOAC, another flat panel. So, again, I'm at my time, but just some takeaway points is, just using standard fluoro could give you a lot of information, valuable information for your nodular biopsies. I mean, if you could take your fluoro and you could see something on your fluoro and you could spin it into angles, you might not even need a robotic scope or imaging technology. You could just use a catheter and a radial probe. If you kind of know where you need to go, you could isocenter properly and you do a good spin and understand, you know, the directionality of what you're doing. Take advantages of the features of DT, you know, these things could overcome CT body divergence. They could provide, you know, the confidence of tool and lesion, which is really what we're hoping for, right? To me, that's a success. A tool and a lesion is a success. The biopsy is secondary, right? There's primary endpoints, tool and lesion, secondary endpoints, you know, yield. And it could be done in your standard bronchoscopy suite with standard fluoro systems that are, you know, are sometimes super old. Yeah, now it comes. Pairing DT with other technologies such as robotic scope. So, you know, very frequently I'll use my ion robot with body vision, you know, it kind of combines the best of both worlds. You could do this. NOAA can put everything together into one system. So, again, these are, DT could be, you know, a useful adjunct to these technologies as well. And remember, image quality is contingent on several things, including, you know, type of fluoro that you're using, the detector itself, and the quality of the recon algorithm, and getting things into ISO center. So with that, I'll end. Thank you very much. So we're going to move on from the Prius to the Model 3 right now. So Brian Husta is the Outpatient Director of Interventional Pulmonology at MSK. And he completed his IP training at the combined Harvard program. And he will be talking about portable imaging. Thank you. Well, thank you all for the opportunity to speak with you all. I hope everyone's enjoying their time here in Hawaii. I've been tasked to speak about the 3D revolution. It's a new, exciting field. And it's great that I'm following Dr. Sasinia. And I'll be adding, mostly talking about mobile cone beam CT. You'll hear a bit more about fixed cone beam CT. I am the Medical Director of our Endovroncular Therapies Program, which essentially encompasses our robotic bronchoscopy program at MSK, which we've had since 2019. And we've had mobile cone beam imaging since 2020. So we've had quite a bit of experience with that. You can see my disclosures listed down here at the bottom. So what I hope to have you all come away from this talk with, understanding the importance of portable imaging and the definition of 3D imaging. You've heard some of that already from Dr. Sasinia. So I'd like to follow up some of those points. And then we can talk a little bit of how mobile imaging is being used in peripheral bronchoscopy. Why is this important for us? Well, we all know too well in this room that bronchoscopy has been one step behind diagnostic bronchoscopy, percutaneous biopsy by IR. You can see the diagnostic yields listed here. We're getting closer, but not totally there. What I'd like to point out is some of the differences between these procedures. In bronchoscopy, it's essentially catheters that are floating in airways versus a percutaneous biopsy where they can push that needle right through the lesion. You've all probably been there where you can see the lesion and you're putting your needle and your catheter is being pushed away with whatever tool you're using. It's very frustrating. We have multiple different tools as listed here. One of the exciting ones is the addition of cryobiopsy, which I think many of us have really embraced in our practice, whereas the percutaneous approach, they have FNA, but also those core needles, which we kind of are envious of sometimes. But maybe with cryo, we'll have that edge over them. Is that difference now down to imaging, that difference between our diagnostic yields? Well, maybe. And you've seen that we've already listed some of those technologies here. I'm going to talk mostly about cone beam, but you've heard about tomosynthesis already. IR is using these multi-slice CTs, you know, diagnostic CT scans. They have great imaging, much more radiation. But they really are able to show in three-dimensional space what they are. I'll draw your attention to this image on the right. This is a mobile cone beam CT image. You can see the frontal. You can't see my arrow, but top left is frontal, sagittal on the top right, and bottom left is axial. And you can pretty much make out where that needle is on three planes. I would say this is 3D imaging, true 3D imaging, because you see in all three axes, X, Y, and Z, where your nodule is and where the tool is. And it's measured. Now, it definitely has greater detail compared to fluoroscopy, right? You saw some of those points just made. The ability to see tool and lesion, I think, is very important, as was just highlighted. But I do want to highlight some of the differences between tomosynthesis and tomography. Tomography is measuring the digital images, and I have a slide kind of showing that going around the patient. Tomosynthesis tries to do a lot with little, right, with just a 60-degree sweep, as Dr. Sisinium mentioned. And is that image good enough? I think we are learning more and more about that as people gain more experience with it. But in the true definition of things, when I was researching this topic, three-dimensional imaging in the radiology language, if you take that from the tomosynthesis that is used for breast imaging, they don't specifically call tomosynthesis true 3D imaging. Yes, you can get X, Y, Z information, but you have to keep in mind that it is using some data that is not actually measured. It is using an AI algorithm to kind of fill in those gaps. So again, applaud the fact that they are trying to do a lot with a little, a little bit of information, small amount of radiation dose. But I'm not sure that it's true 3D imaging. I'll give you an example. When we all came here, we were probably finding this convention center on our phone, and you used three satellites to triangulate your position here in latitude and longitude, right? It didn't tell you what floor you're on. You didn't know that you had to come to the third floor of the convention center. If you want three-dimensional imaging, like an airplane requires three-dimensional positioning to identify your altitude, you need at least four satellites. So there's something to be said about having additional images, and I think that we are going to be learning more as a field. Where is that line? Where is the tradeoff between, you know, is tomosynthesis enough to get you the diagnosis, or do we need for certain lesions true tomography to really confirm our tool and lesion in that location? So some of these general principles have been presented already, but what I do want to show that the beam, the source, you heard about the flat panel that kind of detects the image on the other side. But in cone beam CT, the image that is sent out is actually in a cone fashion, and it rotates around the patient. And you can see that there is some level of scatter that can happen with that, whereas IR, when they're doing their percutaneous biopsy, they have this fan multi-slice CT. Those images are then collected, reconstructed to develop a 3D volumetric image, and you can see what a cone beam CT looks like here on the left and a multi-slice CT. Comparing these two cone beam CT technologies, there's a difference between mobile and fixed. I've been tasked to talk to you about the mobile side of things, and you can kind of see the fixed and the mobile up against each other here. Fixed system here on the right, mobile on the left. They're obviously mobile. Cone beam CT is mobile. The radiation doses are different. The time for the scan differs as well. The time for a scan in a mobile CT scan has less power. It takes a little bit more time. The augmented fluoroscopy that was mentioned available with tomosynthesis has also a feature with fixed cone beam CT, which is that virtual image that pops up after you acquired the images of the full scan. Now, there is a feature that's available within robotic-assisted bronchoscopy that can update the target based off an intra-procedural scan and tell you, tell the robot where that nodule is based off of that real-time scan, hopefully overcoming some CT to body divertance. That is something we're actively studying with the confirmed trial. The cost is significantly different, not only for the machine itself, but the fact that you have a room that you may not be able to flex and use for other fields, for other services. And the image, I think, is rather acceptable. The question of radiation comes up all the time. When you compare the radiation dose between mobile cone beam CT and fluoroscopy, this essentially does the math for you. You have 200 projections and a 30-second spin. It comes out to about 7 frames per second. And in fluoroscopy, if you use 7 frames per second, this is equivalent to half a minute of fluoro time. Many of us are very familiar with that unit. But there are some practical aspects of mobile cone beam CT that I want you to take away from this talk. It is very important to understand that you cannot image the entire chest with this technology. You have to center the area that you're going to biopsy in AP and lateral projection at the beginning of the procedure, especially if you're going to use a sort of robotic-assisted bronchoscopy. You don't want to have to re-register everything, find out that you are not imaging the nodule. You have to use a compatible bed, perform this usually under a breath hold, and consider deploying the biopsy tool at that time. There are different scan qualities, and you can see them listed here. Lower quality will have less radiation dose, and you have higher quality scans with more radiation and more time, which you might want to use in larger patients. This is what it looks like. Once you navigate with a robotic bronchoscope out to a target, deploy your needle and spin to acquire those cone beam images, and it will then reconstruct on its platform to give you that information of where you are in three-dimensional space, and this is what it looks like. I'm trying to walk you through what a case actually looks like here, and you can see that even though we are very close to the target, we are not there. In terms of medial lateral, we're kind of facing towards it, but if you look at the sagittal view, if you deploy your needle, you're going to miss it. You're going to be just below, and if you look at the frontal plane here, you just don't even see the nodule in the same plane as your catheter, even though you are facing that virtual target. So you're not there, even though the robot is kind of telling you that you are, and the way we overcome this is using that information to then adjust the catheter. We knew we had to move into that anterior space. Now we can deploy the target and confirm in three dimension that we are actually in the target, and it's very much like playing darts. I did mention to you that we can update the target with mobile cone beam. After acquiring that spin, you can identify where the lesion is, and then this information will be sent over to the robot to update your virtual target. You can see us here in three planes lining up the target, identifying where that catheter is as well, so then you can just kind of overcome that divergence that happened. Confirming that we are aligned, that a catheter is aligned, and you can see where the new updated target is, and then that is right on the robotic platform and can be updated. What data is out there? We're still in the early phases, but I think it's encouraging because there are lesions, two lesion that is being achieved at a high rate, very low pneumothorax rate, so I think it's rather safe procedures you can see listed here, and I do want to end with a slide that kind of summarizes some of these different topics. Now this is comparing x-ray imaging during bronchoscopy. You can see listed in this orange box are all the mobile platforms that you can use. You've seen fluoroscopy, C-arm tomosynthesis, AI plus C-arm tomosynthesis, and mobile cone beam CT, which I've mentioned about, fixed, and then what IR has all the way on the right as kind of our gold standard trying to get to that, you know, bridge that gap with imaging, kind of catch up with them. The left column here shows the difference between the detector quality, the image acquisition, right, which is that range where you kind of sweep around that nodule to obtain the image. This is where I think it's very important to identify that this is kind of that GPS, you know, satellite imaging that you're getting, being able to identify the image of the target itself, and the ability to identify three-dimensional tool in lesion, radiation doses as you can see here. So going through these, you can see that, you know, the range of rotation around the nodule between C-arm tomosynthesis and AI with C-arm tomosynthesis is using, is only about 40 to 60 degrees. So to image that entire nodule, you may not be getting all those different points, but is it good enough? I think that's something that we are learning more and more about. Taking that step further, of course, now you're trading off some, you know, better image quality. You could maybe be a little more aggressive with your biopsy, but you're imaging a bit more. And, yeah. And you have the ability to reconstruct the images as well. And you can see that different comparison. This is my last slide, actually. When you're comparing the 3D tool in lesion, the breast imaging literature shows that the 3D tool in lesion is kind of called pseudo three dimension. But is that good enough? I think that we still need to figure that one out. But I'll leave you with this, that the comparison of all these different, you know, tools in 3D imaging and bronchoscopy is exciting. Things that we really didn't have before, but now we're learning more and more about it. Combining it with robotic bronchoscopy, I think has been a great thing. And I will leave you with that. Enjoy. Thank you. So, now we're going to move on to the lucid air. So, Abdullah Al-Reyes, or better known as AB, is an interventional pulmonologist, director of interventional pulmonology at Advocate Health. He trained at Cleveland Clinic. He's one of the most dynamic speakers and entertaining people I know. And I think this is going to be a great talk. Thank you for the invitation. And actually, when I got the invitation from Chris and Russell, there was like, I can't find their title to match the presentation that I'm going to go over. And just, just wait. It's not, it's doing the same like yesterday. Yeah. Out of range. Yeah, now it's trying to save file. I think. Crisis averted. Yeah, so I am really lucky to be part of this faculty and be with one of my mentors, too. And really, this is not saying that this is what you should do, but I was lucky enough to have the opportunity to have a fixed comb beam that the staff in the hospital doesn't like to use it. So they said it's available if you want it. And that's what ended with. And I'm going to really highlight more what now we can do with it, not saying that every case should be done with that, not saying that this is the only thing that we should do using, but maybe that some selective difficult cases can be the way we can use it in the future. And I have only consulting with intuitive that has nothing to do with the talk about radiology topic like this today. And I will go, most of the topic about radiation covered, but really the modalities that we have now, we have the digital thermosynthesis portable, and then now we have also the fixed comb beam that can give us more advanced imaging, number one. And two, ability to manipulate around the lesions, the catheters, the tools, to make sure really more accuracy happening when we are trying to get a biopsy from that. And this, everybody aware of this study that we know that robotic by itself is not enough to get us to 90, 95% from even the preliminary results that shows 20% still has to be covered by another imaging to get us that percentage of yield that when we need to get to that number. And fixed comb beam is really the idea of it has this 3D imaging that's giving the options for us to get that as highlighted by Dr. Sinha and Dr. Hosta that the comb beam that rotates around the patient and give us that frequent imaging ending with the 3D imaging, but also the detector size on that comb beam is different from the others. It's larger. It's between 30 to 40, the other 30 by 30. Also, the spin is faster and that's also reduced. I notice it more in cases when you have patients with lesions around the heart because with even the heart palpitating sometimes with those lesions with a longer spin that goes for 30 seconds, you will have some artifact of the motion of the catheter. So, that's a big advantage when you have cases that really adjust next to the pericardium. And the cuts is .6 millimeter cuts and has those 390 images when you do it. The ability also to segment the lesions with the advancement also now with the new CAT scans that available not only lung suite, there is now smart CT. You can segment the lesion. You can segment the airways, which is give you landmarks to drive to. And also, it give you also the anatomic structures that help us in such cases later. I will show you how to prevent even complications. And with that, it allow us to have 3D, not only augmented lesion, but also tools if we need to in difficult cases. And that told us, we learned as the team I work with, the flow of the case changed for us. Usually, we navigate, we reach the lesion, we spin to find it, and then we confirm we are there with any other tool, either with another spin or radial ebus, and then we biopsy. What we learn now to make the flow faster and get other faster way to navigate to the lesion, actually, we spin first. We spin first before we start the case. And then the patient already centralized, and we created an algorithm that prevent us from missing a point where we have to sit back. So we went with ABCD here, as you see. A is anesthesia, so every case, even if we don't have the same anesthesiologist, we go over what the settings we need to prevent better actuses. The bed height, very important in any modalities we are using because you know that sometimes if we don't get the right height of the bed, then we try to centralize the lesion, and if the lesion posterior, and you need to go down now with the arm that maximally down, you will miss that spin. So that's what we learned from our mistakes. That's why we get that B. And C, the clearance and centralization, which is very important for any modality using the 3D, and then we dock the robot, and we go for diagnosis. So the anesthesia, we now need to go over the details, but everybody knows that this is what we, to prevent that electuses, we go over high P, high tidal volume, lower FIO2, and then we go for the bed height. What we learned that the best height that you go with the three feet, that will be giving you a good height for any spinning tool that you're using with radiology with any case. And then C, centralization and clearance, very important to remember when it's really the patient's higher than you have room for your C arm to move up and down, that will give you the option to not miss the posterior lesions. And also, the centralization to put the patient on the table, because that's the disadvantage of using robotic bronchoscopy. If you dock the catheter, you can't move the patient anymore, which will make you delay the case. That's why we went on those algorithms to prevent any mistake that set up back to prolong the procedure. And then the spin, as you see here, it takes really five seconds. Five seconds, hold breath, it's done. And then from that, we go, and this is an example of a case that we spin, and it has been anterior pericardium in the right middle lobe, as you see in those areas here. And after the spin, we do, before we dock the station, we do segmentation. So we segment the lesion. As you see, while we're segmenting, it's creating on the 3D here. After segmenting the lesion, we move to really finding our airway and find, we call them breadcrumbs, put them one by one to get the best airway that we see, right, life in the case, to get to the real, not only time imaging for the lesion, but also real-time airway that will give us advantage while we are doing fluoroscopy to know that we are navigating with the plan that we have in the robot, but also at the same time with the fluoro that we see during the case. When we finish putting the beads, as you see here, we have them in different angles. Then we confirm it with different modalities, and we can even look it up and understand where our ROI or if we need to use lateral more than anterior, PA, or what angle we use, we decide based on this setting. And then we extract the other anatomy, and we're now ready to go for the case. At that point, we dock for diagnosis, and we navigate. We navigate under two images, the fluoro. While we are driving there, we can see the fluoro also live. We go with pulsating every now and then. This is not real, this overlaying the video of the robot over the augmented, just to give you an idea how they are synchronizing together. I hope in the future we get something like this. We have the robot augmenting on the fluoro with the lesion, so we have better understanding where we're heading with that. So as you see here, every now and then when we hit one bread crumb or one marker, we move to another. When we reach down farther, and that's where the deviation happens sometimes, and that's where one airway can take you away from the lesion versus another. You can see here, it's telling us that airway to take, but on the fluoro, it's taking us away. So we adjust based on the fluoro augmentation, and we move toward the lesion. In this particular case, because it's next to the pericardium, you can see when we placed the radial, we got first signal going toward the pericardium, and now you can see the lesion with the pericardium pulsating next to it at that point. So for this particular case, which is not every case we have to do, we re-spin again. So now the catheter, as you see, is really at the level of the nodule. But to avoid that point that you don't want to biopsy the pericardium versus the lesion, you can, with this technology also, use the visal. That's usually for the vascular lesion, people use it. We highlight the catheter as a visal, and now we have it augmented even on the fluoroscopy itself, and that allow us to adjust based on that and get to the right lesion. And based on that now, we can see our biopsies done under not only vision of the lesion, also augmented catheter at the same time. And how we use that to troubleshoot other stuff, because atelactasis itself, even by implementing that protocol I told you, Dr. Casal has this great study that shows that really we are getting fake results from atelactasis. And we know from the strategy using preventing atelactasis, which implemented with our protocol, we still get patients who has atelactasis. And this is an example of a patient that has this nodule that lighting up, but also when we did the spin, we faced that. We got atelactasis. Now, how we can really know more than sitting and counting the ribs and make that our landmark, that's how we use it. The second thing we try to recruit and getting that segmented lesion there based on that finding. And after we segment it also, we start struggling finding the airway. But here, an advantage of lung recruitment, when we have like we asked the anesthesiologist to have really big, 20 feet for 30 seconds hold, the central atelactasis is worse, but the peripheral one improve. And that allow, in those cases also, to be doing the same thing, segmenting again the lesion and also segmenting the catheter, allowing us now to be exactly on the lesion. The second thing that if you have a tool out of your catheter like this, and this one was the radial tip, you can do also segmenting only for that particular part. And you can see, you pick the color you want. And the reason we are picking always orange and green, that was the best produced on the augmented fluoro when we tried. And that confirmed even now with the radial that we have the segment, and now we move to be central more after adjusting that. Deviation troubleshooting, we of course, that's one of the things that we were looking for. And this is an example of also a case of six millimeter in the upper lobe that usually also has this, when we advance to that lesion, usually we reach it. And we have the, just we'll see here how the augmented fluoro helped us to understand the PA and lateral. And when you swim, that's what Dr. Sesenia highlighted earlier, the importance of seeing that lesion on another dimension. The AP and the lateral, you know in the AP you are on the lesion, but on the lateral, you are really away from it. And the adjustment done based on that, and with that adjustment based on really the landmarks of that you got to the lesion in those cases. The last one is very quick case that the patient has a blip next to the lesion, not only highlighting the lesion, but also we can highlight the blip as a target that we don't want to go to. And that helped us to prevent a lot of cases where there's a lot of COVD changes. This patient was needing it because she's a BLVR candidate, but she has this nodule need to be sampled to rule out cancer. And that how you get that use of this technology, not only to go to the target, to avoid damages that can lead to a pneumothorax. And with that, I think with the real-time imaging is very important tool for us now. With this advantage, as I say, it's not particular for every case, but there will be cases that needed for this technology that IR can do, the patients having that large blips. At the same time, this can be the future for accuracy for therapeutic options. Thank you. So, I am truly honored to introduce our next speaker, Roberto Casal, who's a professor of medicine at MD Anderson. He's not only one of the most gifted bronchoscopists in the world, but he's also a brilliant scientist and one of the people I've always wanted to emulate a career after, a kind mentor. Thank you so much, Russell, for your kind words, and thank you, Chris, and everybody for having me here. It's a pleasure to be here today, so let me get my presentation here. So I'm not going to be talking about any of these fancy imaging modalities, but I'm going to be trying to give you some information that's going to help you preventing atelectasis from ruining your party when you're using all these modalities with all your fancy tools. So these are my disclosures, but again, I'm not talking about any technology, so none of these have influenced my talk, and this is more or less what I'm going to go over. So as these guys mentioned, in the past several decades, we struggled to break that barrier that we had in diagnostic yield and peripheral bronchoscopy of the 50 to 70 percent, and I think that the research that we did in the last couple of decades particularly have helped us revolutionize the field of peripheral bronchoscopy is the CT or image guidance and robotics. Now we can get further, and we have a tool that allows us to see where we are, right? So now with all these fancy techniques, we can go after small nodules like this one that are against the pleural, no airway leading to them, but when we put them to sleep and we navigate to a lesion and we try to put the needle out, we see this. Not even his segmentation is going to help you with that, right? Because it's a dense, consolidated, atelectatic, you know, lobe, so that's it. It's game over, right? So that's why I'm going to try to help you prevent that. So when we think about atelectasis and why they occur, you know, if you just think about the transpulmonary pressure, which is the gradient that you need to keep the alveoli inflated, if you keep that formula in your head, you will understand and you will figure out why atelectasis is happening, right? So the transpulmonary pressure is the alveolar pressure minus the pleural pressure. So anything that decreases your alveolar pressure or that increases your pleural pressure will, you know, tend to develop atelectasis. And this has been studied by the anesthesiologists, because this is not new to the surgery. In the surgery literature, the anesthesiologists have very well, you know, studied this. And one of the main things that happens is when you put somebody under general anesthesia and you paralyze them, the muscle tone decreases, the diaphragm goes up, all these viscera of the abdomen put pressure and increase the pleural pressure, and the weight of the lungs as well, you know, will create atelectasis in the dependent area. So there's different, multiple different hypotheses, there are factors of dysfunction, the alveolar gases, I think that the, just the compression and the mechanical compression of the weight of the medicinal structures and the lungs and all the abdominal viscera are pushing up. Those are the main reasons why you will get this. And when you think about bronchoscopy, we have other factors, in addition to the patient related factors, right? Some can do much, you know, obesity, ascites, pleurofusions. But then there's stuff that we do during the bronchoscopy that what it does is it decreases the alveolar pressure. So every time we start suctioning, or if we are obstructing like more than 80% of the lumen of our ET tube with a thick bronchoscope, we inflate the balloon, and they bleed when we take, we're in the major airways, in the central airways, it makes it even worse, right? It's not a peripheral bronchoscopy, this is much more pro-atelectatic than a peripheral bronchoscopy with a thin catheter in the, out in the lung. So all these things, you know, but we can avoid them, right? We have to do them, it's part of what we do. And then there are factors, you know, that could be related to the way we ventilate our patients. And that's what I kind of, you know, try to, in the past, you know, that's what we could start working on to prevent this. So how often does this occur? Well, so the first report, you know, this has been well known in the surgical population, but in bronchoscopy, until we started using CAT scans, we didn't realize, you know, that we were getting all these atelectasis. And this was the first report. We did a pilot study on Combin CT, and we followed these four patients, and we said, we thought we were on the lesion, the radioprobe shows something solid. But then when we got a CAT scan, it was actually atelectasis. So I said, well, this looks very important to me. So maybe, I mean, if you're doing, if you use a bronchoscopy radioprobe, it can give you false positive images and a non-diagnostic sample. If you're using a CAT scan guide, it's still, if it's dense enough, it can still obscure your lesion, and you won't be able to take it. And if you're using any navigational bronchoscopy, it's going to increase. It's not the only factor. I don't think it's the major factor. The major factor is movement, but it's going to increase your CT to body divergence. So I said, well, we have to study this properly and see how often it happens. And that's why we did the iLOCATE trial. And we looked, you know, in patients undergoing bronchoscopy and general anesthesia, the majority were EVAs. You know, we looked with the radioprobe ultrasound in all the dependent areas, and we looked for either like an aerated pattern, as you see in A, and all the other ones are non-aerated. And again, all these images are all atelectasis. These patients did not have any lung nodules or masses in those areas, and they look just, you know, like they can clearly mimic a tumor. So we found, you know, we did this atelectasis survey. The average median was 33 minutes of general anesthesia. That was the characteristics of the population. All these were ventilated with 100% of I2 and nothing or up to like five of PEEP. And at least more than 50%, you know, of the posteriorly located segments of the lower lobes were atelectatic when we looked with the radioprobe EVAs. And the two main factors, not surprisingly, were higher BMI, the higher the BMI, the more atelectatic segments, or they're longer the time under general anesthesia. So that was really impactful, you know, in our field. Because we always thought, like you can see that algorithm there, right? So we always try to think, you know, that when we have a high suspicion for cancer, we used to think that, you know, we would just go to the EVAs. If one of the lymph nodes was positive, we're done. But if they're negative, then we do the peripheral at the end of the case. But not knowing, you know, that, you know, time is very important and patients are getting atelectasis, well, that may have challenged, you know, the way we practice. So how do we prevent this? So there's two types of strategies to me, you know, there's the ventilatory strategy and the positional strategy. So the first thing that we studied, you know, as soon as we were done with the ILK trial, was this VESPA trial, the ventilatory strategy to prevent atelectasis. This was a multicenter randomized control trial. Our primary objective was to determine if this strategy could prevent atelectasis, you know, after what we call the time two, you know, after a long period of anesthesia, because our idea was to still try to be able to do the EVAs first, and then, you know, do the robotic bronchoscopy or any peripheral bronchoscopy, right? So it was a one-to-one randomization. We did stratify them by BMI, knowing that that was the most important risk factor for general and for atelectasis. And we have a blinded review of the scans, and this was done in two institutions. So the VESPA, what the VESPA consisted on, it was instead of having an LMA, we intubated patients. We used a tidal volume of 6 to 8 mLs per kilogram of radial body weight. We did a recruitment maneuver right after we intubated them, and then if at any moment there was any disconnection of the circuit, we did again the recruitment maneuver. PEEP only 8 to 10. Again, the closing pressure of the alveoli, it's 6 in general, 6 to 8. So you only need to be just about that, you know, to keep the alveoli open. And we titrated down the FIA2. We randomized them. As soon as, you know, we intubated or put the LMA, we did the first survey with a CAT scan, and we used a radial probe as well. Then we bronched for 20, 30 minutes, and then we did the second one. So we used the same definition of atelectasis for radial probe, and we used for CAT scan, it had to be really dense atelectasis, 2 centimeters at least from the starting of the posterior wall. That was the definition for it. We had 38 patients in each group. And so if you can look at the population, the mean BMI was 29, which is actually the mean BMI of the U.S. population. So this is representative of the population. And highlighting the time too, there was 40-something minutes, 40, 43 minutes, because that's, you know, where we did our primary outcome, where we looked at the primary outcome and for how long, you know, VESPA was able to prevent atelectasis. So when we looked at any atelectasis, either unilateral or bilateral, we were able to decrease, you can see from the control group of 84% to 28.9%. If you look at bilateral, from 71% to 8% or 7.9%, and if you look at the progression from 76% to 20%. So but it did not go to zero, right? And so this is the same, we looked with the radial probe, it was about the same. There were no complications and no difference in complications in either groups, you know, in both groups, it was like 25% to 27% of hypotension requiring transient use of vasopressors. But there was no difference between the control group and the VESPA group. So again, this was the first, this is still the only randomized trial. We thought that, you know, VESPA was well-tolerated and it has a sustained effect, you know, despite doing the EVAs in between for 40 minutes, you know, and still was able to reduce this, but not get it down to zero. So there are, there is another strategy that has been published, there's two publications on the lung navigation, ventilation protocol, and the last one, they added a very long breath hold as well. So I'm here highlighting the differences, these are retrospective single centers. But the most important thing is they all looked at the primary outcome was the rate of atelectasis. And if you can look in green at the bottom, atelectasis, we all got about the same, right? Like a quarter, less than a third, you know, a little bit more than a quarter of the patients, despite anything that we did, you know, they still develop atelectasis. But if you look at the red numbers, so using twice as much tidal volume and twice as much PEEP, we got the same result. And all of your patients, like 90% actually got hypotensive, 70% requiring vasopressors. So I would recommend against doing something that gives you the same outcome, the same rate of atelectasis, but puts your patient at risk, you know, right? So there was no report, this is retrospective, no reports of a stroke or a heart attack, but no anesthesiologist is going to be happy if 70 to 90% of your patients are going to be hypotensive. And we shouldn't do that because we know that just keeping the PEEP barely above the closing pressure does the same trick, it's the same result, right? I mean, unless in the future somebody does a comparative trial, for now we can say, I mean, you're getting further benefit. And in fact, you're using twice as much tidal volume, which is going to just ventilate anybody. When you're in the Bronx suite, turn the flue on, put in 400 cc of tidal volume, and look at the diaphragm, and then put in 800, and look at the diaphragm, it's going to even disappear from your field of view, you're going to have to move down your C-arm, right? So that is shooting yourself in your foot, because you're causing a lot more CT to bite divergence and movement of your target. So in one of these two in the later publication, they also added like a four to six minute breath hold, you know, which please don't do that, you know, it should be studied in a trial, you know, with entitled CO2 monitoring, that was even not even described. So please don't do that. But so exactly, you make everything move, even in the upper loss move, because of a high tidal volume, and then you need to do an apnea of six minutes to take a biopsy. That's not the way to go. So for now, we can only, from the scientific standpoint, we can only recommend one ventilator strategy. But as you can see, as I said, again, no matter what you use, a quarter of your patients will get atelectasis. Those are typically the ones with very high BMI. And none of these people, you know, is going to get rid of that. And that's why, you know, we thought as we were doing the VESPA trial, what else can we do? Because we were still seeing patients developing atelectasis, despite our VESPA strategy. So we say, how about if you put them on the site, you know, with the lung where the lesion is located, put the lung up, see if they get atelectasis or not. This may be important, I mean, for these very large patients, and when the PEEP doesn't do anything, or when they get hypotensive, we do use of PEEP or recruitment maneuvers. And how about if you put them on the site and just ventilate them with normal, like low PEEP and high FI2, and what happens? So this is how we did it. We put the patient on the site with a beanbag, as if we're doing a thoracoscopy. It's a, for us, you know, with this robot that we use, when we went in, you can see the carina that looks perpendicular to the virtual airway, the actual carina. So you just rotate that, and then you work as if the patient was in supine. So I'm going to go fast now. So we get zero atelectasis, it's just 11 patients, and it took us a while at the beginning to put the patients on the site, now it takes us 10, 15 minutes. And now we're doing a randomized trial of VESPA versus LATs for patients with posteriorly located lesions. And in fact, we enrolled now, like, 13, 14 patients, and we had to, it's a crossover design. We had to crossover three already from VESPA to LATs. VESPA, they got atelectasis, you put them on the site, you recruit them, and the atelectasis go away. And just like this, it's the only, every time, you know, in the past when we had these dense atelectasis, we could back them, we could try to do anything, and it wouldn't get rid of them. But if you put them on the site, and then you do that maneuver, recruitment maneuver, we have found that the atelectasis resolve, and then you can see what you're doing. So in summary, just keep in mind your transpulmonary gradient formula, because that's what's going to help you identify what you're doing that may cause more atelectasis. Atelectasis are very, very common. They can truly ruin your procedure if they become dense, like the ones that I show you. They can give you a false positive radioprovivus image, or increase the CT2D divergence. And I think that preventing is very important. In some cases, you will have somebody who has a lymph node that looks like it could be positive, and you may want to do the EVAS before doing your peripheral, because that may be it. So it's good to have a strategy that you can use, you know, to try to prevent it while you're doing that. I think that so far with the evidence that we have, maybe in the future there will be comparative studies of VESPA versus something else. I don't think, I would not do that myself. But for now, we can only recommend VESPA as an advantageous strategy. And the problem is that it doesn't work with all the technologies. I've tried with other technologies, other than the robot that I was using, and it could be a little bit impractical or difficult. I mean, there's ways to adapt and to do it, but it may not be as easy. And then, whatever you do, also, just keep in mind of the high tidal volumes. It's not high PEEP. The high PEEP is what you need. High tidal volumes will move everything. I mean, you will do a pulse, but even when you're navigating, you don't want things, you know, to be going up and down. If you have a lower lobe lesion and you use 10 mLs of tidal volume, that's going to move all over the place. I don't recommend that, I don't think. In fact, what we're doing is I'm dropping the tidal volume. I'm using the PEEP, you know, but actually, if it's a lower lobe lesion, I use a lesser tidal volume. And that's it. So, thank you so much. So, we've run out of time, but the next session is ours, so we will have a few minutes for questions. Just before I get to that, this is the second of four ABIP sessions at CHESS. We have two more coming up, one on having a successful procedural service, and then a great one which will definitively decide if RIDGID or FLEX is the king in bronchoscopy, which occurs at 3 o'clock. Please rate these sessions. We want to be able to continue to give ABIP talks at CHESS next year, and the feedback that we get is going to matter for our ability to be able to do that. Thank you. ♪

imaging modalities

bronchoscopy

atelectasis prevention

standard bronchoscopy systems

digital tomosynthesis

cone beam CT

mobile cone beam CT

3D imaging

augmented fluoroscopy

ventilatory strategies