Diane Newman: Welcome. I'm Diane Newman, I'm a nurse practitioner, an Adjunct Professor of Urology and Surgery at the University of Pennsylvania in Philadelphia, and I'm here today to introduce to you Dr. Anne Agur that I heard at the SUFU meeting a couple weeks ago. She presented some basic science research on the area of neuroanatomical approach to neuromodulation. Dr. Agur is a Professor in the Division of Anatomy, Department of Surgery at the University of Toronto. She's been a teacher and researcher in the division for more than 40 years, with primary research interest in clinically applied normal versus pathologic structure and function of the musculoskeletal system. So, please join me in welcoming Dr. Agur.

Anne Agur: Thank you very much. I'm very happy to be able to share some of our findings and the way we collect data to develop novel models that look exactly like the nerve in the specimen. The models of the human saphenous nerve that I'm going to talk about will be used to develop neuromodulation therapy approaches, both an engineering finite element approach, and for clinicians to look at electrode placement.

To get high-fidelity data, which hasn't been possible until now, we've developed a technique in our laboratory that involves dissect, digitization, and 3D modeling. The first square shows dissection of a specimen. You'll see the full dissection shortly. You can see there's a tiny nerve, and all of these are branches of the saphenous nerve, with one of the branches that is being dissected out in yellow and the other in blue. What we do is we take a digitizer and you digitize points along the nerve. We go at 1 millimeter increments. What this does is it takes 3D Cartesian coordinates of all the points that we touch, whether they're bone, nerve, anything we want to preserve to use this landmarks. You're probably wondering, why is the screw here? We have three screws, and these are used as reference markers so that we can put the data together when you do a 3D model.

This is what the specimen looks like. As we're following along the screen on digitization, and these are all programs we've developed over 20 years and they're very unique to our laboratory for the data collection, we can see these points, and those are actually indicating the points we digitized. And you can see they're joined by lines so we can see what we're doing. And if we're off, we can delete a point.

Then we go to 3D modeling, where, again, we have had a huge emphasis on developing the programming. So we can put this together in Maya using our home-grown add-ons, and you can see here is a digitized nerve. We usually indicate them as cylinders. These are pretty big because then we can see them well right now. And what are they placed on? They're placed on a laser scan of exactly the same specimen. So the idea is, is that you are able to reproduce what you see in 3D and you can turn it and twist it any way you like. You can make a simulation out of it so that it's hugely advantageous. The big problem with finite element modeling to date of procedures and including saphenous nerve modulation to treat overactive bladder syndrome, is the lack of high-fidelity data.

3D data. Usually people have used simple dissection. Once you've opened the specimen, you can't do any more than that. The nerve will get shifted, cut, and you make photographs and you just have 2D representations. So what happens here is, this is a fully dissected leg from the knee to the medial malleolus at the ankle, and you can see branches of the saphenous nerve in the various colors. Furthermore, these branches have not been followed in detail in decades. In the early part of the century, there's a little bit more description about them. You can also see the big vein, the great saphenous vein here, which could also be a landmark. So that in the middle picture, then, is the 3D model, only from the medial view. Obviously we can't show a movie and twist it and turn it here on the 2D slide. You can see the outline of the leg, the way it was. You can see some of the bony landmarks, many more than what we've labeled, but the medial malleolus and patella are here, tibial tuberosity is here. So you can see the various branches and the great saphenous vein.

So, if you look at where's the greatest concentration of nerves, well, right around the knee, but we could also follow these branches and look at where you could get good electrode placement to stim the nerve, so that you are just not looking for it in a rather unorganized approach. So we have to remember, we could use ultrasound guidance to insert a device. If we look at the next picture, this is showing the tibia, and this is the bone that we digitized. We digitized way more when all was said and done, but this is the critical piece for this nerve.

We can see all of the surfaces and the various features of the tibia, the patellar ligament, the tibia tuberosity, and you can relate the position of these nerves. You can mathematically measure in 3D space and you can then devise protocols, potentially, to look at where the device should be placed, the neuromodulation device, which is a huge advantage. And then using this approach, you could also needle where you think it is and lead the needle in and see if you're anywhere close to where you think you were supposed to have been. So the efficacy can be tested.

But how this project really started was with [inaudible] and looking at developing high-fidelity finite element models to actually simulate nerve conduction following stimulation. So, this shows how we can get very high-fidelity data that's never really been possible before, and it's 20 years of work, but it's a very specific Cartesian coordinate data.

So what is our take home message? The 3D data provides a very precise in situ map of the distribution of the saphenous nerve for nerve stimulation, for developing overactive bladder syndrome treatment protocols and to test these protocols. It also provides high-fidelity 3D data to facilitate more accurate finite element modeling of the whole stimulation process for neuromodulation, so that in the anatomy lab, we work on a multidisciplinary team and it is all about collaboration and getting all these pieces together. And it's a hugely clinical collaboration, as well as one with engineers and other computer scientists. Thank you.

Diane Newman: Well, thank you. This is really fascinating. How difficult was it, though, to identify the saphenous nerve? Because I see that you're right, there's so many nerves within the lower leg, or mid to lower leg, and they do move and they may not be exactly the same place in every leg. But that was your first step, to identify it in an actual leg.

Anne Agur: We do many, many, many, many legs.

Diane Newman: Do you really? Yeah.

Anne Agur: And you won't capture all variations, but if you get the main patterns, which we seem to have pretty well, sometimes you will have unexplained findings, but you can then start to develop what you think would be good protocols. Where do you place the electrode? Where do you place the implantable device? Where may there be a good place? Saphenous nerve gets tricky because it's a sensory nerve, which we want. It's only sensation. You would rather stimulate only a sensory nerve, not one that's mixed and will make your muscles fire too.

Diane Newman: Right.

Anne Agur: So that you really want the sensory nerve. And this is a very new approach to really try to help people that have this quite common problem.

Diane Newman: Well, yeah. You bring up that the saphenous is only sensory. We are stimulating the posterior tibial nerve, and that's both motor and sensory. And you're right. It's nice that this nerve is specifically to sensory and a lot of what the OA reactive biosymptoms are, of course, are sensory with urgency and frequency now. So 3D is being used more and more, isn't it, for research such as this?

Anne Agur: Yes.

Diane Newman: This is the unbelievable technology to be able to do.

Anne Agur: Yes, 3D is being used a lot more, but in our lab, we have a very unique protocol that isn't being used.

Diane Newman: Really good.

Anne Agur: And what this also enables is, if we look at augmented and virtual reality, when you have 3D Cartesian coordinate data like this, you can make all sorts of educational simulation tools using the data. It takes a lot of time and it takes effort, but it's so high-fidelity that it will mimic the positioning of the tissues very, very well.

Diane Newman: Is there a long time, though, before you can convert to that 3D, or not really?

Anne Agur: No, we're doing it right now

Diane Newman: Really, huh?

Anne Agur: Yeah.

Diane Newman: And you always start with actually looking, say, within whatever part of the muscular cells, and then you move into the 3D technology.

Anne Agur: You can do muscles, you can do everything. In this case, we're looking at the saphenous nerve for a very specific reason. It takes a long time to digitize.

Diane Newman: Does it?

Anne Agur: Yes. You can imagine all the points you need to get and you have to do it so that you don't disturb the nerve, the muscle, whatever you're looking at. It has to maintain its position in the specimen, and we're able to do that very well.

Diane Newman: Well, thank you so much. This was so fascinating. I know when I listened to your abstract at SUFU, I found it really... It's nice to know the background, because we kind of, as clinicians, always know the end result. Now we have this new treatment. But I know there's many years before that, and really, a lot of individuals involved in this, so thank you very much.

Anne Agur: Thank you.