Dual Energy CT in Emergency Radiology

Thank you for the contradiction and again thanks for Siemens for inviting me to present here and also this very inspiring meeting. I want to talk about dual entity. We heard a lot about to Lefty already I want to kind of highlight a couple of applications that we've used to improve our clinical performance. So we're kind of translating initial research results and actually changing our protocols in routine to improve our daily routine work. Never less, if you heard from Brazil for Glaceau our true ER. Cases are scanned on a sliding entry city, so when when I talk about emergency dual entity here, it's going to be more about vascular pathologies and some MSK in there and. What we've seen also a thing from all the vendors there is a great interest in doing ICT. Certainly every vendor has a concept for it. Think were true. Difference lies as the post processing. I mean all these scanners can run do lengthy. But the true benefit of density is really when you analyze this data and change the image perception an ultimately transfer these images to PACs. One thing I really want to clear out at the very beginning, 'cause I get this every time and also in reviews. There's it's those neutral. So just two papers for this. When we run a scanning tool and she and at the bottom it's going to show you. Sorry this is both for the flash scanner and this is going to be for four scanner, both optimized single energy scans we found nonsignificant differences so that's why we're undoing city in routine work in all patients. That's going to be all vascular. It's going to be all abdominal scans. And also, analogical scans of course. You've seen this slide before. Also, the reason that we run these VM eyes so this monolithic imaging as they were trying to improve the image quality and especially investor doing ICT you have to go drastically lower from the energy level so you end up close to cage. So you want to end up mostly at 4050 KV and that shows you also that if you there's a difference really between trying to achieve this goal with locate VP scanning and just doing it. We are past processing on doing ICT. And there's very initial work that was done for the Manheim Group, where they showed with the initial algorithm and Stephanie possible to improve the vascular enhancement when you go to very low energy levels but the same time. With this algorithm, there was very high in noise, so that's why this technique was very rarely used. Nevertheless, there are certain even perspective studies with this initial algorithm showing we can use this for contrast reduction. Thankfully, there was a great improvement in this technique. You've seen this before from the Cameron showing that with the newer algorithm on this, the true difference is the decrease in image noise, and that's where we start to get more interested in and really use it in clinical routine. So what you can expect this is the same scan. Again, this is all contrast enhanced just at very high windowing settings and showing you the same drastic increase investor enhancement but at the same time much lower image noise. Adjacent to it, just using this we call it noise optimized via my. We don't like menu plus as a title and just for vascular. This is the true difference compared to what your standard dual energy scan will look like. And as usually when you try to adopt these techniques at the very beginning we just look at image quality and we obviously we saw you've seen this last summer before we can dress, increase image quality with this technique and our next steps were by other groups. Shop is for basically every vascular. Territory to look at our next steps were to what can we make of this in clinical routine? What I really like about dual Angie is compared to when you try to do like a 70KV scan where you have one shot and if it goes wrong you have to repeat the scan with dual energy, become much more robust and also much more flexible and this is good example because for example with the VMI technique, if you want to increase vessel contrast you can do that at very low energy levels, But if you want to for example you have patients with stands or extensive classifications, you become much more flexible because you can just alter the. Energy level without repeating any scans and you can become better at imaging these standard vessels. And here's an example just from a paper we done. There were on the blended images. The readers were kinda not sure if this is really stenosis and got unproved on the VMI and confirmed on calf. So there was kind of translation in our research also to go away from just looking at image quality and say, well, it's probably better to really see can we improve our performance? And this was one of the earlier papers we did on this where we really saw a difference just by this process processing and becoming better at looking at vastus kenosis of the lower leg lower extremities. We've seen it also from other groups that really you see a true difference in diagnostic performance or increase just by this post processing. I think this is important. This is where auto clinicians got more interested in this and this is where we're actually right now. Routinely reconstructing these images when we run these scans. Emergency setting there's a couple of opportunities for this purpose. What we'd really like to do is an acute bleeding because bile intravesical contrast, of course, will be increased. Also, extravasation will show a higher contrast, and here's example were on the centered lending you see in this patient with partial spectrum E and affected me. You see, some the bleeding site here, but it's really drastically increased here, so more subtle bleeding size will be more enhanced just by processing. At the same time, we've heard this before. We also like to use this technique because it allows us to save radiation in VNC for certain opportunities, especially definitely in emergency setting. So showing this just for abdominal bleeding we saw an increase just by this process and technique where we were able to not only be more more sure in the diagnosis, but also an increase in the diagnostic performance of actually detecting abdominal bleeding. He heard a couple of more examples showing what these images look like on blended the old algorithm and then the noise of the most algorithm where you can go much lower from the energy setting. What you see. Also, I want to emphasize this when you use these techniques because the contrast is so much higher, you have to alter your Windows settings. So obviously this is not how you would look at standard blended image and you send it setting, never less. It shows you the drastic difference. So as a next step. Even when, especially when it's very subtle bleeding, you become much better. He's another patient after a vaster access, where the bleeding size is somewhat there. But it's really much more appreciated on Postprocessed via my images. Again, high window settings, so I want to show you this, because to get a fair understanding of this technique, you really have to alter your way. You look at these images, which makes it makes it more complicated again. So at Standard Vester Windows settings, you see this very high signal that is actually. Too much, so you really have to alter this and there's a great paper from from the Charleston from Joseph where doctor check your shoulders first for hepatic lesions that if you use these post processing techniques you have to enter your Windows settings, but then they become really efficient and we looked at this also for. Diagnosis of pulmonary embolism were of it. This is an easy case, obviously, where there's a huge embolus here. This is how you would look at these postpositively and I image is if you look at them with standard Windows settings. But if you use optimized Windows settings and we've calculated actually a conversion factor for this, that can be used routinely for all cases, then demarcation of these emboli becomes much more easy. What we've also heard already today and what we also use routinely, is our profusion Maps. There is one difference, I think, for us that we like to use them as an addendum so we don't fully rely on them, but we want to use them in edge in connection with themeid images. And when we looked at this in the study, so there are three protocols that we separately looked at and one is just your standard drag ECT. Scan blended image. Second is you have to scan. And you have to perfusion map and the third one is you have a scan perfusion map an via my image series and we really saw a difference in diagnostic performance for diagnosing PE and this is obviously especially peripheral and very subtle PE and also an increase in overall confidence by the readers, particularly also for segmental PE. So another case that I think showcases this a little bit more. There's a subtle embolus here, much more much better visible on the via my image and in connection with the perfusion map. It all fits together and you get more certainty for diagnosis and the true benefit. I think that we've seen is that now these techniques which were in the very early research stage can now be routinely calculated so you don't actually have to sit on the single via for all these. Images again instead of the scanners, sends the command to the remote server and the image lens and in your packs. Because we is radiologists, although I know seems like this to be different, we don't want to set a single all the time, especially in emergency settings. So we want to look at these images on packs because we have big screen etc which is much more efficient for us. And I think there's a true difference to other concepts of dual entity were ultimately the images end up in your packs automatically and there was a huge difference for us and that's where we routinely now. Have you my plus images in every dual energy scan and then a piece scan? We will also have these perfusion Maps. And there are very interesting study from the colleagues into Bing. I think where they showed this is comparing oncological patients that underwent routine staging city. So this is not at all an angiographic case and there is no history of symptoms etc. But nevertheless there looking at incidental findings of PE later on. Confirmed by see TPA and they found just by using these VMI reconstructions, lot of Emily detected on. Just staging CT, there were not really visible. When looking at these images and your standard scan, maybe because I'm not particularly looking for it, but on V. My plus. It's really the contrast is so much better that you're the diagnostic accuracy becomes better at the same time. And refused this technique also. Now for other sites were vascular injury is present and for us this was also a lot of for endoleak because analex complicated topic probably needs multiface. But again if you can increase the Vester signal and also the signal there shouldn't be you become much better at your diagnosis. I want to briefly look at some MSK applications that we heard about MSK V4. We've all seen these images were with the older algorithm. It's a good correlation, especially in patients. Also, product process that you can visualize the bone marrow edema. I think this works particularly well also because sometimes you lose track, so you might be too focused on these fractures here. There are obvious, but actually in this study one of the two of the readers. Missed this more subtle fracture just because they were so focused on where the fracture site is and where the findings are more obvious, so I think it's a technique to improve your performance because you're gaining more information from this whole scan. And this is just an overview of a paper where we looked at readers for different experience, different levels of experience, and we saw that all readers saw an improvement using these Palmero Maps some more, some less. But this was actually the most experienced reader in this study from the time, and so substantial increase in his diagnostic performance, because, again, you get more security in your diagnosis. So this is a new algorithm and the difference not only is visual for this with the new Siemens, but it's also a difference, because now you can automatically get these images if you have a good collaboration manager on site and it looks different but at the same time it for us it open the opportunity for other applications of this, so we're not only looking at bone marrow edema of the spine. This is a very subtle under non displaced fracture here that probably you might have missed on here unless you really looking for it. And so I think there's a big future for Ballmer edema or get in general edema visualization using doing HCT. Also in the thing, see T examinations were usually will do MRI. But nevertheless I think if you have this indication for CTU gain gain so much more information from this just by using post processing. What we've also seen there's a great revenue article on do Lanja applications in MSK. They're looking at vertebral disc herniation and we've used these. This approach also, but this is also the beauty of doing ICT. The ways to get the visualization or kind of multiple ways to get there, so this is the Visitation parent that they used. For example, we switched it up because we thought we can improve the visualization of the herniated discs and just kind of change up the colors to get better. Here, of course there are multiple very obvious herniations, but I think for example in this case it's a little bit more subtle, and looking at your standard grayscale scan think you become much and our data supported. Also, you become much more sincere in your diagnosis and you have an easier visualization of these injuries that might be present that you would probably miss in routine, so maybe similar to my presenters before. I'm also suggesting to use more density because you gain more. Nation with no penalty dose. There are certain limitations that Doctor Marion showed this morning. I think there is a lot image series still getting reconstructed, so it might be confusing for you in packs. But if we can simplify this and increase time to get you on packs you will gain so much more information from standard scans that you might have missed before. So kind of summing this up I think especially for contrast enhanced applications and specially for vascular deal. See T doing city really gives you benefits over other techniques and not only and that's good. That goes beyond just image quality, but really translates into improved performance and finding more pathologies. Actually and in MSK I think there's a good future for Bar Meridian visualization, not only of the spine but also of the extremities were I think T right now is playing more of a role of planning surgery and it's not really diagnostic where so using more. I think my city might be the way to get there. Thank you for attention.

Virtual Monoenergetic Imaging (VMI) VMI+ Image Quality VMI+ Window Settings Vertebral Disc Herniation VMI+ in Clinical Routine Dual-Energy CT: Principles Dual-Energy CT: Radiatipn Bone Marrow Edema Dual-Energy CT: Radiation VMI+ Diagnostic Accuracy Diagnostic Accuracy in Clinical Routine Image Quality => Accuracy Image Quality in Stents VMI+ 4. Thorax (2nd Gen DSCT) 1 oow f. o. 40 v F 0.5 F_O.5 40-kev VMI+ Dual-Energy CT: Radiatipn 70 10 Dual-Energy CT in emergency radiology Energy CT in emergency radiology o se SIEMENS SIEMENS .. Endoleaks after endovascular aortic aneurysm repair: 1000 improved detection with noise-optimized virtual " Protocol 3 Protocol 3 Julian L. Wichmann, MD Julian Wichmann monoenergetic dual-energy CT SIEMENS .. Julian L. Wichmann, MD SIEMENS university Hospital Frankfurt, 0.4 0.2 1000 Dep. Of Diagnostic and Interventional Radiology, o. (AP in F_o.5 40keV Heol-thineers • Healthineers university Hospital Frankfurt, University Hospital Frankfurt, Frankfurt, Germany 0.2 0.4 0.6 0.2 CNR in o' to CNR in to Schenzle et al, Invest Radiol 2010 Leithner, Wichmann et al, Invest Radiol 2017 Alchecnt, Wchmenn et al, Invest Radiol 2016 Schenzleh et al, Invest Radiol 2010 Wichmann, Schoepf et al, Invest Radiol 2016 Schenzlechman et al, Invest Radiol 2010 Aecht, Wihman et al, Invest Radiol 2016 (P Julian L. Wichmann, MD Results: B-W/L and O-W/L for M70 were 460/140 and 450/140, and were 1100/380 and Department of Diagnostic and Interventional Radiology Department Of Diagnostic and Interventional Radiology the Of Abdomen (2nd & 3rd. Gen DSCT) 40-kev VMI 40-kev VMI+ 70-keV VMI 40-kev VMI+ M 0.6 80 W 1070/380 for M40+, respectively, differing from standard DE-CTPA W/L settings (450/100). 1000 125 75 F_O.S F_O.5 VMI+ VMI• Paranwter (P s , 039) (P S , 039) standard linearly•blended (a) and images virtual photon standard linearly•blended (a) and images with virtual photon Energy Level (keV) 2nd Gen 120.kV tradition* series (Amas 0.992. 0.916, 0.914. series (Areas the 0.992. 0.916, 0.914. 40 10 parameter F _ 0.5 Highest subjective scores were observed for M40+ regarding vascular contrast, embolism SIEMENS of 80 kev 100 kev (c) and kev (d) of an iliac artery stent with a 40-kev VMI+ 40 kev VMI+ 40 70 40 kev VMI+ ! iameter of 6.0 mm show increased lumen visualization and reduced blooming artefacF ! i ameter of 6.0 mm show increased lumen visualization and reduced blooming artefacts Gen IOO-kV 0817 0775 the Stent V With Same the Stent With the Same Gadoliniun Gokj Gold Go'd Bism._'th (0.71-090) k.sne köne (089-0.991 (089-09971 (066-0871 Gen Healthineers demarcation, and overall IQ (all p