- 261
Transcript
This is [inaudible 00:00:01.695]. And why you're looking at this nice guy here in the bottom, one with the thermal camera image and another one with the data camera system from the Dynamic system. I was just saying that the ExacTrac Dynamic system is still a work in progress. And therefore, we have not been using it on patients, but on phantoms alone. And just an overview of my institution and my radiation therapy in Denmark. We're quite a big department with 10 LINACs. Eight of them has ExacTrac, one have the nonclinical Dynamic system installed. Three are 6.5, and two which coming more to be 6.5. And all these are going to be upgraded to a Dynamic. We have one MR LINAC at the moment. And then Copenhagen is divided in two departments Halo [SP], and Richard Taylor [SP]. And we have low staff in the orange area. And we are also serving the radiotherapy for Greenland and Faroe Islands, yes, you heard. We are also serving radiotherapy for Greenland which was quite famous in the newspaper when Trump wanted to buy the... Sorry.
So ExacTrac Dynamic is a new system where you integrate, combine the surface scanner system together with the ExacTrac x-ray systems. And this is, you can see, the right side of the image is what you used to see, but now you also have the left-sided image where we have the surface camera sitting in the middle of the room. And the camera system is...integrated system with a thermal camera in the bottom, a structured light projector in the middle sending out blue light, and then two data structured cameras. And this is maybe 4D structure which has a frame rate of 20 megahertz. I think that is quite high compared to what the Visionati and the CRET system is, running more or less the half of that frame rate, and a latency of 400 megahertz. And the two images, the heated signature and the 3D reconstruction is then put together to a 4D surface, where every point have an thermal value together with a 3D structure.
Why is this of benefit? Well, problem is if you just have 3D structure and a flat area and you want to track that, for instance, for gauging, you need some more information sometimes because as you see on the bottom right side, you have just a very...area where it's very difficult to see if something is sliding. But if you combine the 4D thermal information we can definitely spot where we have a robust surface, which we can track much more precise. And then the system should be, as we're told, less sensitive to skin and light in the room. This is just a single surface scanner system. And you might sense that if you compare to the Visionati or CRET, they normally have three cameras. So of course, we have less 3D information in our setup. But I don't think that is the main drawback because now we have an integrated system where we have both the x-rays and the surface camera integrators. So we know, those two systems knows what is happening with the positioning of the x-rays and the surface. Of course, with less in the roof and also less sensitive against to calibration because you don't have three structured cameras trying to get the same surface. And also think this is giving the chance of a higher frame rate of the system.
This year, I just promised BrainLab to show this. This is their new Brain Fixation system port. For me, it's most important that now the system has a very fine 4Pi clearance. It's possible to have a double-face, double-mask system. And now it's also possible to have an open mask because we now have the surface scanner system tracking the patient all the time. I think the open mask is interesting for claustrophobic patients but also for the patient comfort. So this is something, I think, can be nicefor a lot of people. Of course, there's new calibration phantoms. Of course, now we have thermal and structural camera systems to calibrate. So this is the first, the example of images we get and combine. And in the bottom, we see the calibration method for the x-rays. And here we use now the onboard imaging system and the radiation isocenter for the calibration of the ExacTrac system. And here's the phantom we've been using for the testing, and cranial phantom, and a heated body phantom where we can heat different kind of areas with different kind of temperatures so we get a heat map on that phantom too. And the surface camera field of view is quite large. It's 60 by 47 and also 47 in height. So this is quite a large area. We can use the surface information. And then we have the x-ray field. In the top, we see the ExacTrac system images. Here you have 30 by 30 field of view. The flat panel is 30 by 30, and the DDR here, and the image, when you try to match the field, it's the same size, 30 by 30 square centimeters. The new ExacTrac Dynamic has a larger field of view of x-rays, so it's 18 x 18. The flat panel is still the same size. But the DDR is the whole DDR from the planning CT. I think that's very nice features because now it's much more easy to see how to...when you reconstruct you want to ensure that you are matching correct. It's also easy to understand what you are seeing here in the images. So I think this is a nice features and I think also when you go to all sites, the larger field of view is much more interesting and ensuring a safer workflow.
And now we come to the workflow. Oh, that's nice. No, it disappeared again, sorry. Now let's set up software. The running workflow is now a little bit different. We have a pre-setup with a thermal camera, surface camera, where we combine the information from the CT scan, the planning CT and the system now tries to move the patient to the correct position by that information. So that is moving the pre-setup of the patient. And after that, we go to the... We can go and define the area of interest, if it has necessary aisle [SP] for the tracking area or for the open surface mask, we can just want to see what we would like to see.
Then we go to the positioning. And that is done as we do an ExacTrac with two x-ray images and that is compared with what you have for the planning. But at the same time, the system also takes a reference surface. And that is something I think is quite well because normally when you have a surface scanner system you have to tell it when you want to make a new reference. And you never know when you can take it before you make the match. And that is some kind of time delay in that way. But now here, this is just automatically and the surface system knows now if you move the patient how much you're moving the patient. And then, of course, you're going to treatment mode and here you can monitor the patient with the surface monitoring all the time and also triggered by x-rays. You can trigger by surface motion, or by mounting units, or gantry angle.
Coming small movies now to show that. And so let's start with the setup movie. So here we can see how the system is trying to move the patient in a correct position. And when it does that, it also tell us which part of the surface is not in correct position. So we can maybe help to set up the patient better. After that, we can do surface area of interest and with a brush probably define where we want to, in the surface, to track the patient. We also have seen already that the air temperature of the nose change when the patient breathes. So we now maybe have to be taking a little bit care about that. But we have not really tested as in clinical use yet. Then, of course, we're going for the positioning and referencing. And here you know all the items as you normally do. There are also some new items, which is the rubber band, like it'll just come in a moment.
Here is the rubber band, you just pick the...pull one of the images very short, and then you just go back as the rubber band. And I think that is quite easy, understandable, but it's quite effective in seeing how the patient is lying correctly. And then, of course, we recommend all to take a reference, newer reference, new precision verification image when you have moved the patient. And then we come to the monitoring. And this is quite a PC slice when we see the movie. So then take the first to explain what you can see in the next movie. On the right side, top, you see the thermal image of the patient, where you have some bars showing how much the deviation is. And then the top, the total distance of the difference. In the middle of the bottom, you see the time during the treatment how much the surface is changing. And you can say there tolerance line of...the thin white line there is the tolerance. And if the tolerance is o-run, then you can go to the machine, the ExacTrac system, it automatically stop the beam. On the left-top, you see the x-rays and they can be acquired out by the...in a fixed beam per monitor units or when you make a rematch plan you can do it by gantry angle. And now the ExacTrac system knows where the gantry is and knows which x-rays is able to take an image. So if it's high, one of the ExacTrac tube is high by the gantry, you just dose and take that one. The system is analyzing the monitoring x-ray images alphabetically, and that takes about one or two seconds and then you have an answer there. You can either decide to continue with the treatments or you can wait and analyze, wait till the system has analyzed the two new images. You can also...when the surface scanner is going to be changed or out of tolerance, then you can also acquire new images of x-rays. And you can also pause and get the repositioning of the patient and then continue the treatment.
So here we have the style of the treatment. And you can see how the patient is lying quite in the same place all the time. Now we acquire some images, it's analyzed, and its tolerance. Now the patient is moved and the system is stopping out [inaudible 00:13:49.420] and the patient is coming back quite soon and then you just continue, or you can wait and take some extra images. So I think this is a very nice tool and I think that is what we're thinking, improve the treatment of brains here because we now have a much more safe way to deliver the treatment. We have also been testing the accuracy of the ExacTrac system and compared to what have been found earlier. And if you see that the three first lines of the table is from literature and our data is at the bottom on phantom. And we can see we get the same precise of ExacTrac Dynamic as we see from other people from the ExacTrac system. We also been trying to see how precise is the surface scanner, surface compared to the ExacTrac x-rays. And they are more or less equal. Of course, this is was on a phantom, so we expect that we're a little bit different if we have a real patient, but that's promising at least.
So in conclusion, we have a new system with ExacTrac Dynamic. We have a larger field of view compared to the ExacTrac system. The surface monitoring with data and thermal camera is integrated. The surface monitor integrated with the x-ray imaging is pre-positioned with the surface camera. Improved treatment verification, monitoring during treatment, surface: all the time, x-rays: gantry angle and monitoring triggered. We have a fully integrated system with the LINAC. And the software seems very fast and easy to use. And then we believe we are already working with BrainLab about that. There's going to be some kind of deep inspiration breath hold possibilities with the new ExacTrac. The first version is without that software, but the next versions is coming later. What I understand from BrainLab is that they're expecting that the [inaudible 00:16:06.375] ready end of year, start of next year. And the deep breath hold software should come approximately nine months later. So at the end of the next year same time as now or maybe a little bit later. And of course, if you can go down and make breath hold techniques, we can have all indications. And of course, if the higher value of the system because now more patients can be used with the ExacTrac Dynamic. And also the open mask is now something I think is interesting for a lot of patients. And I won't go into this breath hold because that is not...it's still so much work in progress. So that's what the perspective is, very nice.
So I think just to end my talk, I think BrainLab on prostate is very nice with ExacTrac because it is even better with ExacTrac Dynamic because now we have the monitoring during treatment and we can do with the open mask. I think we can go from spine and spine compression patients because we have a monitoring during treatment but also a larger field of view which makes it much more safe to verify the patient positioning. With the breath hold software, I think we go for breast, deep breath hold breast treatments, and also the bigger field of view is also necessary for that kind of treatments. And then I think lymphomas in mediastinum, lung and liver, in combination with CBCT and soft tissue match and deep inspiration breath hold would be part where we also would use the ExacTrac Dynamic, and maybe esophagus. In our institution, the best lung, and lymphomas, and liver is standard techniques and breath hold techniques. So, to stop my talk, thank you very much, here's my team. And here you can see how big the field of view is on the ExacTrac Dynamic.
So ExacTrac Dynamic is a new system where you integrate, combine the surface scanner system together with the ExacTrac x-ray systems. And this is, you can see, the right side of the image is what you used to see, but now you also have the left-sided image where we have the surface camera sitting in the middle of the room. And the camera system is...integrated system with a thermal camera in the bottom, a structured light projector in the middle sending out blue light, and then two data structured cameras. And this is maybe 4D structure which has a frame rate of 20 megahertz. I think that is quite high compared to what the Visionati and the CRET system is, running more or less the half of that frame rate, and a latency of 400 megahertz. And the two images, the heated signature and the 3D reconstruction is then put together to a 4D surface, where every point have an thermal value together with a 3D structure.
Why is this of benefit? Well, problem is if you just have 3D structure and a flat area and you want to track that, for instance, for gauging, you need some more information sometimes because as you see on the bottom right side, you have just a very...area where it's very difficult to see if something is sliding. But if you combine the 4D thermal information we can definitely spot where we have a robust surface, which we can track much more precise. And then the system should be, as we're told, less sensitive to skin and light in the room. This is just a single surface scanner system. And you might sense that if you compare to the Visionati or CRET, they normally have three cameras. So of course, we have less 3D information in our setup. But I don't think that is the main drawback because now we have an integrated system where we have both the x-rays and the surface camera integrators. So we know, those two systems knows what is happening with the positioning of the x-rays and the surface. Of course, with less in the roof and also less sensitive against to calibration because you don't have three structured cameras trying to get the same surface. And also think this is giving the chance of a higher frame rate of the system.
This year, I just promised BrainLab to show this. This is their new Brain Fixation system port. For me, it's most important that now the system has a very fine 4Pi clearance. It's possible to have a double-face, double-mask system. And now it's also possible to have an open mask because we now have the surface scanner system tracking the patient all the time. I think the open mask is interesting for claustrophobic patients but also for the patient comfort. So this is something, I think, can be nicefor a lot of people. Of course, there's new calibration phantoms. Of course, now we have thermal and structural camera systems to calibrate. So this is the first, the example of images we get and combine. And in the bottom, we see the calibration method for the x-rays. And here we use now the onboard imaging system and the radiation isocenter for the calibration of the ExacTrac system. And here's the phantom we've been using for the testing, and cranial phantom, and a heated body phantom where we can heat different kind of areas with different kind of temperatures so we get a heat map on that phantom too. And the surface camera field of view is quite large. It's 60 by 47 and also 47 in height. So this is quite a large area. We can use the surface information. And then we have the x-ray field. In the top, we see the ExacTrac system images. Here you have 30 by 30 field of view. The flat panel is 30 by 30, and the DDR here, and the image, when you try to match the field, it's the same size, 30 by 30 square centimeters. The new ExacTrac Dynamic has a larger field of view of x-rays, so it's 18 x 18. The flat panel is still the same size. But the DDR is the whole DDR from the planning CT. I think that's very nice features because now it's much more easy to see how to...when you reconstruct you want to ensure that you are matching correct. It's also easy to understand what you are seeing here in the images. So I think this is a nice features and I think also when you go to all sites, the larger field of view is much more interesting and ensuring a safer workflow.
And now we come to the workflow. Oh, that's nice. No, it disappeared again, sorry. Now let's set up software. The running workflow is now a little bit different. We have a pre-setup with a thermal camera, surface camera, where we combine the information from the CT scan, the planning CT and the system now tries to move the patient to the correct position by that information. So that is moving the pre-setup of the patient. And after that, we go to the... We can go and define the area of interest, if it has necessary aisle [SP] for the tracking area or for the open surface mask, we can just want to see what we would like to see.
Then we go to the positioning. And that is done as we do an ExacTrac with two x-ray images and that is compared with what you have for the planning. But at the same time, the system also takes a reference surface. And that is something I think is quite well because normally when you have a surface scanner system you have to tell it when you want to make a new reference. And you never know when you can take it before you make the match. And that is some kind of time delay in that way. But now here, this is just automatically and the surface system knows now if you move the patient how much you're moving the patient. And then, of course, you're going to treatment mode and here you can monitor the patient with the surface monitoring all the time and also triggered by x-rays. You can trigger by surface motion, or by mounting units, or gantry angle.
Coming small movies now to show that. And so let's start with the setup movie. So here we can see how the system is trying to move the patient in a correct position. And when it does that, it also tell us which part of the surface is not in correct position. So we can maybe help to set up the patient better. After that, we can do surface area of interest and with a brush probably define where we want to, in the surface, to track the patient. We also have seen already that the air temperature of the nose change when the patient breathes. So we now maybe have to be taking a little bit care about that. But we have not really tested as in clinical use yet. Then, of course, we're going for the positioning and referencing. And here you know all the items as you normally do. There are also some new items, which is the rubber band, like it'll just come in a moment.
Here is the rubber band, you just pick the...pull one of the images very short, and then you just go back as the rubber band. And I think that is quite easy, understandable, but it's quite effective in seeing how the patient is lying correctly. And then, of course, we recommend all to take a reference, newer reference, new precision verification image when you have moved the patient. And then we come to the monitoring. And this is quite a PC slice when we see the movie. So then take the first to explain what you can see in the next movie. On the right side, top, you see the thermal image of the patient, where you have some bars showing how much the deviation is. And then the top, the total distance of the difference. In the middle of the bottom, you see the time during the treatment how much the surface is changing. And you can say there tolerance line of...the thin white line there is the tolerance. And if the tolerance is o-run, then you can go to the machine, the ExacTrac system, it automatically stop the beam. On the left-top, you see the x-rays and they can be acquired out by the...in a fixed beam per monitor units or when you make a rematch plan you can do it by gantry angle. And now the ExacTrac system knows where the gantry is and knows which x-rays is able to take an image. So if it's high, one of the ExacTrac tube is high by the gantry, you just dose and take that one. The system is analyzing the monitoring x-ray images alphabetically, and that takes about one or two seconds and then you have an answer there. You can either decide to continue with the treatments or you can wait and analyze, wait till the system has analyzed the two new images. You can also...when the surface scanner is going to be changed or out of tolerance, then you can also acquire new images of x-rays. And you can also pause and get the repositioning of the patient and then continue the treatment.
So here we have the style of the treatment. And you can see how the patient is lying quite in the same place all the time. Now we acquire some images, it's analyzed, and its tolerance. Now the patient is moved and the system is stopping out [inaudible 00:13:49.420] and the patient is coming back quite soon and then you just continue, or you can wait and take some extra images. So I think this is a very nice tool and I think that is what we're thinking, improve the treatment of brains here because we now have a much more safe way to deliver the treatment. We have also been testing the accuracy of the ExacTrac system and compared to what have been found earlier. And if you see that the three first lines of the table is from literature and our data is at the bottom on phantom. And we can see we get the same precise of ExacTrac Dynamic as we see from other people from the ExacTrac system. We also been trying to see how precise is the surface scanner, surface compared to the ExacTrac x-rays. And they are more or less equal. Of course, this is was on a phantom, so we expect that we're a little bit different if we have a real patient, but that's promising at least.
So in conclusion, we have a new system with ExacTrac Dynamic. We have a larger field of view compared to the ExacTrac system. The surface monitoring with data and thermal camera is integrated. The surface monitor integrated with the x-ray imaging is pre-positioned with the surface camera. Improved treatment verification, monitoring during treatment, surface: all the time, x-rays: gantry angle and monitoring triggered. We have a fully integrated system with the LINAC. And the software seems very fast and easy to use. And then we believe we are already working with BrainLab about that. There's going to be some kind of deep inspiration breath hold possibilities with the new ExacTrac. The first version is without that software, but the next versions is coming later. What I understand from BrainLab is that they're expecting that the [inaudible 00:16:06.375] ready end of year, start of next year. And the deep breath hold software should come approximately nine months later. So at the end of the next year same time as now or maybe a little bit later. And of course, if you can go down and make breath hold techniques, we can have all indications. And of course, if the higher value of the system because now more patients can be used with the ExacTrac Dynamic. And also the open mask is now something I think is interesting for a lot of patients. And I won't go into this breath hold because that is not...it's still so much work in progress. So that's what the perspective is, very nice.
So I think just to end my talk, I think BrainLab on prostate is very nice with ExacTrac because it is even better with ExacTrac Dynamic because now we have the monitoring during treatment and we can do with the open mask. I think we can go from spine and spine compression patients because we have a monitoring during treatment but also a larger field of view which makes it much more safe to verify the patient positioning. With the breath hold software, I think we go for breast, deep breath hold breast treatments, and also the bigger field of view is also necessary for that kind of treatments. And then I think lymphomas in mediastinum, lung and liver, in combination with CBCT and soft tissue match and deep inspiration breath hold would be part where we also would use the ExacTrac Dynamic, and maybe esophagus. In our institution, the best lung, and lymphomas, and liver is standard techniques and breath hold techniques. So, to stop my talk, thank you very much, here's my team. And here you can see how big the field of view is on the ExacTrac Dynamic.