 So while he's getting set up, next up we have Dr. Kim, one of our cornea fellows, who will be joining faculty at UT Houston next year, and he was telling me he's excited to have his own fellow switch of roles. I know this is a project that Dr. Mifflin gave me before I started fellowship. Remember reading Dr. Smith who's here, his IRB over May of my third year. And something that I've done over the last year. This is my fourth iteration of this presentation. The first one everyone here has heard was at clinical faculty day. And there are some repeat slides, but about 80% of this is new. And it's kind of, I've been adding new things every few months that we do more things in clinics. So as I'm leaving here, I wanted to go over some of the things that I found that we can do with the confocal microscope, just so that everyone's familiar with its potential. And also to go over some of the results of the study that I introduced back in September and to show you the results of that. So these are our slides I've already presented. So they're basics. The confocal microscope basically gives you 3D histology in real time. The benefits are you don't need to take a biopsy, and you can image multiple areas without any trauma to the patient. There's three kind of confocals. Basically as the light gets more coherent, it goes from white to a single nanometer length color, basically the resolution becomes better. The first one there is not used anymore. The slit scanning we have in the NIDEC confocal 4, which we use to look at the endothelium. And the Rossock cornea module is the HRT that we use to look at the glaucoma nerve, but it's a cornea adapter to image the cornea with that and basically again the axial resolution. So not the left-right resolution, but the z-direction resolution improves as the light becomes better. What we have here and the things that I've done research with are the Rossock cornea module, HRT3. It basically has a 300 by 300 micron area resolution, and the axial resolution is 4 microns, which is much better than the other two, and that's basically the main difference between the confocals. It doesn't mean a lot to see it all in words, but if you look at the pictures here, the NIDEC is over to the left, which we have here, and the Rossock cornea module is over to the right. You can just see that every layer of the cornea is just a lot cleaner. You can see you start from the epithelium, the nerves, the stroma, and then the endothelium over there. So this is a sample patient of someone I've done recently, and this is the screen that you would see. The fear that everyone has is this is direct aplanation technology, and everyone goes, well, if you're looking at the eye and you're aplanating it directly, aren't you going to mess up the surface? Well, if you do it properly, you will require almost zero aplanation. It's supposed to barely touch. The gel is supposed to just lightly interface the surface of the eye, and you get these beautiful images. You basically zero the depth, so you know exactly how deep you are in the cornea, too, and you can kind of follow that along. And to compare and contrast, this is a picture of someone I did in July, and you can see that I am pushing much harder on the eye over here, and that is not the proper way to do it. So if you do it like that, you can cause epidefacts and other issues, but pressing it lightly, there's not too much of an issue. So larger pictures here, again, this is the surface epithelium. You can see that very nice kind of tile layer. You go a little deeper, and you can see the subbasal corneal nerve layer, and you go a little deeper, and you can see the anterior stroma layer. The anterior stroma is in a crisscross pattern for the anterior one-third, whereas in the posterior two-thirds, it's in a parallel lamellar pattern. So when you look at research studies using confocal microscopy, the results are everywhere. For Sjogren syndrome, some people say there's increased nerves. Some people say there's no change. Some people say there's less. For post-traumatic infection, some people say there's a decrease. Some people say there's an increase. And basically for every paper you find saying one thing, you can find another one saying the exact opposite. So whenever you have such a variation in results, you have to wonder why that is. To look into that, I think we need to think about the data in two different ways. There's quantitative data and qualitative data. When you look at qualitative data, you think about infectious keratitis. If you see one fungal element, or if you see 10 fungal elements, it's still fungal keratitis. It doesn't matter how much you see. And but for quantitative, it depends a lot on what cross-section you get. If you get two nerves versus three nerves, that's a big deal. And if you look at different parts of the cornea, you may have basically different amounts of sampling error. And when you look at the data, it seems to be more like that the qualitative data is reproducible, but the quantitative data is not. So we're going to start with some qualitative examples. So at Emory, we always used to harass the first year. So and I know all three of them are here today. We're going to start with Jim, because this is a patient I saw with him on Christmas Day. Is anyone surprised to see on Christmas Day with Jim Bell? And unfortunately, this is not the picture of the patient, because we didn't take a picture on Christmas Day. But I found a representative picture. And the cornea kind of looked like this, but the lesion was actually much smaller. Imagine something even smaller than that in the yellow circle. This is a 24-year-old contact lens user. Standard came in with really bad pain. Was started on Vigamux, got worse, and the optometrist called us in a panic. I started fortified, because I didn't really, I don't know, it's just what I do, I guess. If they were on Q1 hour Vigamux, and they got worse, what do you do? But it never really looked all that bad. And I remember scraping this patient with Jim on Christmas Day. So Jim, what do you think? Since you saw this person in real time too. Which one, right? Yeah. Yeah. Yeah, I mean, the eye didn't look all that bad. Yeah, I agree. The eye looked like this. It was quiet. You look at him, and you're like, man, why did this guy drive two hours coming on Christmas Day? But if he drove in and didn't spend time with his family because he was such bad pain, you gotta think, well, there's something really going on over there. So I thought it was fun, low-caratide. It's more of that spicule look on the outside. But I wanted to make sure. So I brought him back on another day, and we did a confocal microscope test. And thank you for Randy for editing this, and slowing this down so people don't vomit, I guess. So this is the surface layer. You can see that the epithelium is intact, but you see some white cells on the surface. And as you go deeper with the confocal, you can kind of see the outline edges of the infiltrate. I mean, there's a distinct edge over there, even though the epi is still intact. And this is me kind of guiding the eye around with the side light, kind of like you have on the Yag or the argon. And if you go even deeper, wait for the picture here, you can see that there's all these spicule structures on the edges. And you see that there's this edematous edge of the infiltrate, and that necrotic center in the middle kind of looks like mush, and there's a lot of white cells in there. And it'll go one layer deeper in a second here. And I just wanted to show you that you can kind of pan across an entire ulcer from one end to the other. And you're going to the other side. So when you look at these individual slides, you get, each of these videos is about 100 photos. And I thought, well, what if I montage it together? And you get a picture of a corneal ulcer. Now this looks gigantic, and it kind of overflowed the confines of the PowerPoint. But really, this is only a half a millimeter ulcer. And if you don't believe me, basically, each of those squares, and I think I ruined the pointer here. There we go. Each of these squares is a 300 by 300 micron block. So if you look at the ulcer from edge to edge, it's really only about one and a half blocks. So that's only a half a millimeter ulcer, but you get so much information from that little area, that little white dot that we saw. So this patient never grew out fungus. I blame you, Jim, for the culture. But anyway, we started on that in my sin, and he got better. So it kind of looks like this, right? The outside edges are kind of spiculed, and it's kind of representative of fungal keratitis. So we're going to do a two-bed Akbar's knot here. We're going to do white dot syndromes now with the residents, but of the cornea. And we're going to go over it with confocal microscope. So Dan, what do you think? This is a similar story. 45-year-old contact lens user was treated with steroids and antibiotics for about two weeks before they got to us. And this is on day two after we started this patient on fortifieds. Yeah, yeah, just a standard stuff, right? I have one question for you. Do you think that's one infection or two infections? Okay. And so this is day three after treatment, and I noticed that there were two white blocks here initially, but this one got better, but this other one did not. And you can kind of see how this one's kind of fading out on the edges, whereas this one's still the solid block. And so again, decided, well, let's get a closer look. And I got to make sure. So I managed to get a video going from one ulcer to the other, right in between, going left to right. So this is a video going from one end to the other. So here you see kind of that soupy bacterial infiltrate that you saw before, and you pan to the other side, and it's very different, right? And you go back to the other one. So I don't know if you guys wanna see that again. You're going from one end, and you know, you don't need to know much about the cornea to know that those two things look kind of different. And that looks a lot like fungal hyphae. And so she has a cone infection. You know, you have a contact lens person who, we had a contact lens user who was on steroids for two weeks. So you have no idea what's going on. This happens to us all the time, like three weeks of treatment with everything under the sun. You come with this like necrotic mask that you have no idea what's going on. And basically she had the bacterial portion was getting better on the fortifies for two, three days, and the fungal portion was not. And you can see it's very distinct. The hyphae elements are on the order of 100 to 200 microns, and you can get very distinct images. So again, this patient we started on Natomycin and Fluconazole, and it ended up the mitre's dose, 200 milligrams BID, took five weeks, but she's finally better. And Leah, what do you think? This is another white dot. She's still like 2100, because it's in the middle. Eventually we're gonna let the scar fade and see how it turns out. And so Leah, so what do you think? This is another white dot, and this is a 45 year old, someone that had a trigeminal nerve decompression on the right. I don't exactly know why you do that. Well anyway, the trigeminal nerve got injured and she has loss of sensation in V1, V2, and V3. We've been treating her for neurotrophic keratitis and she walks in with pain and this kind of white fluffy area. So I don't know, what do you think? Yeah, you're kind of curious, right? And this picture looks so much worse, right? Yeah, it looks bad. And so, yeah, same person. Same picture, just under, sorry? Oh, different lighting, yeah. This is just some Lysclerotic Scatter. So, but this person is neurotrophic. You know, we had the thought, maybe this is Candida, maybe we wanna scrape, but you don't wanna scrape someone with a neurotrophic cornea, someone that we fought with end on end to try to repair their epithelium. And end up destroying the epithelium without knowing what's going on. So we thought, you know what? Let's take a look at what that white material is made of and we'll only scrape if we see any kind of weird elements. So you look and we never saw any kind of infectious element. We didn't even see white cells. You know all those little white dots that we saw everywhere? We did see what looked like very immature epithelium. You see the patient here is on the left and you see a normal patient's epithelium on the right and it's basically this nucleated epithelium in its early stages that was actually causing on that clouding. So we send it to Dr. Patel, send, got a tarsaur fee, patient's better. So a lot of information with the confocal, you get these lot of white dots and you don't really know what to do, but it gives you a better idea of what to do when you see all these white dots and a doctor degree will like this. So again, this is the left eye and this left eye is normal and the right eye, remember, had a trigeminal ganglion, basically injury after surgery. So this is a volume sack. It goes from the top and goes to the bottom. So it's going to go first through the epithelial layer, then through the nerve layer and then through the stroma. And you can see, you see all those lightning bolts there, right? Those are all the nerves. And again, this is a healthy eye over here. So I'm going to replay that one more time and you see all those nerves, right? When you look at the side that has had the longstanding trigeminal injury, there's nothing. I don't see anything over there. And it kind of verifies that she has a neurotrophic cornea and everything makes sense. So that's qualitative data. It gives you an example of why this is useful and why those studies are reproducible. But then you look at quantitative data. Again, how many nerve fibers you have? What are the branching patterns? How many fibers? How squiggly are they? And what's the total nerve density? You can also look at endothelial cell count. And one thing I've been looking at a lot is epithelial cell count. One of the things that we can't do very often is we don't know how to quantitatively grade the health of the ocular surface. I remember when I was interviewing for cornea fellowships, I always ask people, do you guys do limbo stem cell transplants? And I was greeted with laughter about half the time. I think Ed Holland and Dr. Mifflin are the only two people that said that they actually did limbo stem cell transplants. So let's see here. And these are slides that I went through in clinical faculty day. And I'm not gonna dwell on them. Basically, all four of these slides look very different, but they're really the same patient. They're the same actual stack, but they're only separated by one second. And you can see how different each of these pictures look. And if you didn't believe that they're the same person, you can see one common structure in every single photo. And it's just one a little to the left and one a little to the right. And you can see, because the branches have more empty space in between them, you can get a lot of sampling error when you look at quantitative data. And here's some of the reasons why. You know, the pictures that I showed you look beautiful, but they're screened. They're like looking at surgical videos. You know, they never show you the cruddy ones. They always show you the best case scenario one. So this is what an actual patient looks like. Someone that's kind of non-compliant. You can see that not only that you get the X, Y variation, you're getting variation along the Z axis too. This person's coming in and out of the slit lamp. They're moving their eye left and right. And so these images are all over the place. It's actually not surprising when you start doing these that you get this mess of quantitative data because it's very difficult to keep it in one plane. And it's very difficult to go back to the same spot, more because you're aplanating the patient with this gigantic device. And it's just not the most comfortable thing. And patients have trouble sitting in the slit lamp as it is, let alone when you're touching their cornea with a gigantic cap. And even a patient that's really good, you notice that there's still a little, kind of what I call microsecades, but there's still always this little slight panning motion that you get even with all that. So those are some of the variables that kind of affect why the data may or may not be reproducible. The other trouble that I've had is as you go further and further away from the center of the cornea, your images, you start pressing at different, basically you start pressing at different pressures on different parts of the cornea. Again, the cornea's not flat. It is flatter on some parts of the periphery and it's steeper on others. So basically as you're pressing on the cornea, you'll press harder on other parts than others over here. And you can see that on solitary images as you get to the edges, you're in different layers depending on how hard you're pressing. So you see a lot of pictures published in the data and they go, hey look, there is neurotrophic cornea. But you look in the actual pictures in half the wrong layer. And so is it really that you have neurotrophic cornea or is it that you just have the corneal picture in the wrong layer and you don't realize that half of it's in stroma and your numbers are basically half of what they should be. And what I found out is Bowman's layer is not really a solitary layer. It has these undulations and these waves that kind of jut up into the epithelium. And what happens is if you press a little hard with a confocal microscope, it's like those toys that you had, the little pink cushions where you put in your hands and they imprint your hand. You could basically get these corneal bands and they're basically these outpouchings of Bowman's layer. And Bowman's layer really is in that chain-link fence pattern over the entire epithelium. That's why sometimes when people come with very bad dry eye, you see this kind of tortoiseshell mosaic pattern on the epithelium and that's kind of the highlights of Bowman's layer kind of coming up into the epithelium. And all you do is pull back a little and the bands disappear. So what I thought was before we did all these quantitative studies, we got to see if it's even reproducible. It's kind of funny. It's like Dr. Olson's study with the aberrometry. A lot of times when you see numbers, you have this illusion of objectivity. But just because you have numbers doesn't mean it's objective. There are things that are robust like the IOL master where people nitpick when we're off by like 0.01 millimeters and yet we publish hundreds of papers on something where no one's even done simple standard deviation and reproduction from visit to visit. So what I did was we got 10 quote unquote volunteers, 25 to 35, you can see that that's in the range of residents, PhD candidates, and fellows. And we visited, they're all healthy, no medical disease, no eye problems. And I measured them at two different points in time. You know, why are we looking at disease corneas? Why don't we look at normal corneas first at two different points? I made every attempt I could to yoke the eyes together and make sure you're getting to the same spot. That Amzler grid that you see over there is basically separated by two millimeters, five different points. And what I would do is cross cover the eyes and couple the left and the right eye so they're yoked together and use the non-aplanated eye to guide the eye into the positions I wanna be. And if anyone had a correction, I would correct it to have a residual four diopters of correction so that they would not accommodate. And I would keep the Amzler grid 25 centimeters away so that you can always have the eyes in a similar position without that crossing phenomenon as you converge and accommodate. And what I wanted to see was, is this even reproducible? What variable is reproducible? And is there an area on the cornea that's more reproducible than another area? And the way we did this study was, well, we wanted to make sure the images were of proper quality, because a lot of papers you'll see in the literature are not of proper quality. And our proper quality criteria was a 90% of the image was in the same layer. We looked at the nerve density, the fiber count, the branch count and the amount of tortuosity. There's a program that's like a paint function that we trace the nerves manually and they will look at this and basically get these coefficients and measurements. I don't exactly know how to do the tortuosity because the algorithm I'm not too familiar with, but anyway, it doesn't matter. We just want to see if it's reproducible or not. And again, we did it at the center and superotemporal, infrotemporal, superanazel and infranazel. And the reason why we made it in X instead of across was if you made it across, it would kind of get in the way of the actual microscope so we had to turn it 45 degrees so the patient can see all of these four points. All five images were compared individually and the averages of all five were compared. And when you look at an example of this at the nerve layer, you can see that, I mean, it's everywhere. If you look at this guy up here and you look at the center, I mean, the nerve counts look starkly different. So depending on where you look at on the cornea, you can make, I can say whatever I want. You could just mix and match different pictures and go, it's the same, it's less, it's more, just mix and match and make whatever conclusion I feel like. And because you have all these pretty pictures, you're kind of enamored into trusting and believing those numbers without really looking at the reproducibility. And even when you look at two different images from one patient on first visit to second, the fibers are oriented in different directions. So you know you're in a different spot of the cornea. It is really hard. You have an area that you confuse around and you don't always get to the same spot even when you don't have any kind of for you on cross cover. And you have a range that you confuse and you can see that you're in different areas even when you're just trying to measure the center. A lot of the studies go, oh, the center's always reproducible. We'll go back to the center. But clearly from this picture, you can see that they're not in the same place. So when we look at our study, the single image data for all five variables kind of almost doesn't matter what they are because we're just seeing if they're reproducible or not are awful. The error range is 20% to 55%. And when you look at these, if a p-value's 5%, you could have that difference just from sheer variation in a normal person. And this is with every effort to go back to the same spot. Most studies just kind of go back to any spot. But when you average them, you see that all the numbers get better, but they're still pretty awful for the most part except for one. And that's the total narrow fiber density. And so from this study, one, I'm Dr. Mithila, I'm sorry, I'm still trying to write up and we'll get there in a little bit. Basically the only quantitative measure and you could even follow is total narrow fiber density. Because the rest of them, you don't know whether that difference is just from natural variation or from an actual difference. Unless you had such a significant difference and you use these numbers within the statistical calculation to compensate for all that. But a p-value of 0.05 is not really valid to show a difference. So again, we know that basically averaging images gives you more reproducible data but there's still a lot to be done. And really when you look at the literature, the only thing that you really should look at with significance is the total narrow fiber length or total narrow fiber density. So a couple other people do a lot of work with the confocal microscope. When you look at this, this group in New Zealand or Australia I think does a lot with this and this is a postmortem cornea. So you can see that, that's how they got so many images. The person was not really fighting you too much at this stage. And you notice that all the nerves kind of come in from the outside in and they meet at a confluence which they call the vortex. Now you think the vortex is in the center but it's really not. It's located inferior paracentral, right below the middle. And the vortex kind of looks like this. So that's part of the reason why I wanted to measure the center in four other different spots. Maybe you get a more reproducible area if you're going in this area of the vortex but as we saw, that's not really the case. And I really noticed that pattern and it looked really familiar to me. So I opened my cornea atlas and I found a couple other examples where I saw that pattern. And you see that in this Faberase disease patient and this amiodural worlds, you also see that kind of whirling pattern from the outside in meeting paracentrally. The actual disease does not matter. The disease itself is just staining the epithelium and you see the epithelium and how it grows in what pattern. And a lot of people think neurotrophic cornea is not because of the tearing, it's because the nerves are actually the train tracks in which the epithelium grow in. And this would kind of indicate that the epithelium would kind of follow in its tracks a lot better and maybe you're just missing that highway for the epithelium to grow with that. And you can see that this pattern is very similar to that corneal nerve pattern on that postmortem cornea. And when you look at the epithelium, this is the edge of another, I imaged a lot of random patients, a lot of ulcers on top of that. This is at the edge of the ulcer. You can see healthy epithelium, sick epithelium and even sicker epithelium at the edge. And you notice that as the epithelium gets sicker or newer or it's turning over, I don't exactly know where it is in the timeframe because this is a snapshot. The epithelium gets bigger and it's a pattern that I noticed elsewhere too. So let's look at this. Again, these are the epithelium from the neurotrophic patient, one at the edge of the ulcer and one of a healthy patient. You notice that the epithelium goes from being nucleated to anucleated to smaller and denser. And it's something for the first year medical students here. This is why when you decide to specialize you should look into every other field because there's a lot of symmetry in medicine. You look at other cells that turn over a lot and this reminds me a lot of blood cells. They're bigger in the beginning, they're nucleated and they get smaller as they mature. They lose their nuclei and they become denser. And so can you look at the epithelium, see what stage it's in and have these other measure ends of the health of the ocular surface? Well, are these turning over? Are they dying? Are they very healthy? Are they medium healthy? And basically it's like looking at endothelium. You can see endothelial cell count. Are they pleomorphic? Is there polymegathism? You have all these other variables to basically gray the ocular surface instead of going one plus PDK, two plus PDK and doing your shirmers. That's different depending on how much prepare can you put in the eye to figure it out. This is a patient, and these are all real patients that I saw in clinic here this year. These are not out of papers or Google images. This is a patient I saw in Dr. Taven's clinic and Jeff Petty can attest to this. This patient treated in Idaho for an immune keratitis, again treated with steroid for a really long time and the epithelium just looked awful. You looked at it and it kind of looked like a canthamoeba when I looked at it so I decided to image it before I decided to scrape it. And you notice that these, a canthamoeba cysts over here look a lot like the ones you see on a pathologic slide on H&E. And again, this is 300 microns. Bacteria is one to two microns so you really can't see bacteria. And a canthamoeba cyst is 20 microns and fungal elements are 100 to 200 so you see the scale where you should and should not be able to see things. So we decided to do a complete epithelial scrape. We grew this in Pages. It grew floridly positive for a canthamoeba and the patient is now 2020 doing well, work home managing with people with Dr. Cannon in Idaho. And it expects to ask the question, well, how much can you really see with this? Immune cells are in the order of 10 microns. So I don't know if this is for real because people again say everything when you look in the literature or if this is just a Rorschach ink blot test for me where I'm just imagining things. But when you look, it kind of looks like, these are granulocytes and these are lymphocytes and these are maybe eosinophils and this is from the fungal keratitis examples where you see kind of this boomerang pattern of the nucleus. And I don't know if Dr. Mamelis wants to comment or say anything, I'm just imagining things or if this seems fairly legitimate. Okay, got it. And this is a patient I know, Dr. Mamelis is very familiar with. We've sent samples like five different times to, this is that patient with this scleral nodule that keeps on coming back that we've been re-imaging and imaging and imaging. So just like I do with other patients where we have no idea what's going on, I said, well, why don't I do the confocal image of the congenitiva and see what I get? And it's kind of like what Dr. Mamelis describes in a lot of his pathology reports. It's a dense chronic inflammatory infiltrate. But I saw one thing that was kind of interesting. When we've sent this patient's tissue to Seattle for PCR, I sent them back to Dr. Grossencloss in Emery for another analysis. I mean, her tissue's been everywhere. We have no idea what's going on. Maybe I should present that patient next time we have a grand rounds to get some thought. But I don't know, what do you think about this area over here? Does that look like a multi-nucleated giant cell? Is this a granulomanus inflammation? I don't know. I don't know if I'm imagining things again or it kind of looks like that, but yeah. Again, just food for thought. And something very interesting. So this isn't a very inflamed eye. So as I was imaging her congenitiva, I was like, what is all that stuff moving in there? And you can see the blood vessels and the blood cells moving in the congenitival blood vessels. And so here you have the first example. I'm sorry. This patient is in complete down gaze. So there's kind of like a nice stagmus kind of motion trying to go back to primary gaze here. So again, if you get a little motion stick, maybe not watch, and I'll play it a few times. You can see all that blood moving in there. And sorry about that. She's trying to keep on moving up and I'm going, please look down, please look down. And she's kind of shaking a little. And so that's a medium-sized blood vessel. But you can even see smaller ones too. And sorry, I kind of pan around a little to find the area. But you can see all that. All those cells kind of moving in there. So pretty neat. And there's a lot of stuff that you can actually do with this. The one thing it's not very good at is looking at the endothelium though. I've tried a lot. And there's something about the aplenation where you're just bending the cornea in a weird way. You never get it cleanly in a plane. And I think that this confocal with all its potential and with all its great resolution, it's a double-edged sword. The closer you look at something and the smaller your field of view, the more it is to have artifact from compression. And basically, when you have the night egg, you get an image that's about one and a half millimeters. This is 300 microns and I don't know. I've never been able to get a good image. So I don't know. I think that that's one of the limitations. It has beautiful surface images, but I could just never get good endothelium images. And I've tried a lot of times. And so I think that the night egg still has a place. I'm not saying that this is the end all be all for all confocal. And like everything else, everything has its purpose and its own usage. So again, confocal microscopy of the cornea is a very powerful technology, but still has its limitations, especially when you look at numbers. Image averaging is really essential to getting good reproducible data, but even then that's not perfect. And I really think that this is a tool to really look at the ocular surface and to really extend our knowledge. We have a paper on DSEC published about five times every single month, and there's things that are reproduced all the time. But there are things that we can't fix in cornea, like dry eye and neurotrophic cornea. We have no idea what to do with it. I think this has a lot of potential in the future and something that I am looking to concentrate on to try and figure out some of those answers and see what we get. So here are our references. Thank you for having me this year. It's been a wonderful year. Yeah, Dr. Olson. So, excellent presentation. It turns out in the research side we're more of a huge emphasis on... Yeah. I think you can. The problem is that this machine is bulky and you gotta keep that patient in that same position. If they can somehow, almost like a halter kind of, they could kind of shrink this and have it connected to certain spots or even smaller, like almost like a V scan so you can move it around and keep it somewhere. I think it has a lot of potential, but the way it is right now, it might be really hard to have the patient look in an extreme gaze in one direction or the other and center it. But I think as it becomes more portable and more mobile, you can think about doing more stuff like that. But yeah, tons of potential, but a lot of work to be done with it. Dr. Ambari. Really nice touch, Jim. No, it doesn't. I'm using the... That's why I tried to yoke the eyes together and you'd move the other eye, but... Yeah, I think that would be a great issue because you can always go to that same spot. I mean, obviously you have a parallax problem with that just like we do with the refractive lasers, but anything would be better than what we have right now. So there's a lot of ways to do that. I think that's a great idea for volume of the nerves. I have not tried. I had trouble just even doing these 2D images. So I think that's the next step with the image processing. And it seems easy enough to do because there's only four or five microns thick and you should be able to do that and again, actual nerve volume. And I think that's the next step. And I think I just have to... I think I need to compare up with a computer science person to really do something like that. Huh? Okay, that'd be great. Okay, yeah, that'd be awesome. I think that'd be a great next step.