Transcript
Good afternoon. It's my great pleasure to have a chance to make a presentation here in Novalis Circle meeting. And today I'm going to talk about the intra-fractional skull movement within the relevant head and neck mask system for 6D. And first of all, I would like to express my great appreciation to all my colleagues involved in this study. And, we are located in the east part of the Kyoto City in Japan, and this is the hotel where the previous Novalis Circle meeting was held. So, we are very close to that hotel. And this is Imperial Palace, and this is Nijō Castle, maybe some of you visited there at the meeting. And we installed Novalis system in the end of 2006. And from February 2007 we started clinical service for brain and prostate. And in June 2007 we got 6D robotics.
And for frameless stereotactic irradiation, of course, it is very important of the reproducibility of the skull portion. And in addition to that, it is also very important, the intra-fractional movement of the skull should be small. And this high reproducibility was achieved by the ExacTrac X-ray system. And we also validated the overall accuracy of that one. But today, I'm going to concentrate on this intra-fractional movement. So, the purpose of this study was to evaluate the intra-fractional skull movement, fixed in the BrainLAB head and neck mask system. And we used the Novalis and also the ExacTrac X-rays 6D and the BrainLAB head and neck mask system. And we used a CT slice thickness of 1.25 millimeters, with a field of view size 40 by 40 centimeters, and the CT scan was taken with G right speed. And we evaluated the 10 patients for SRS or SRT with this mask system. And what we did was monitor the skull movement with the ExacTrac X-ray. And at first, we localized and corrected the position, by the ExacTrac X-ray. And after that, we verified the positions every 3 minutes, in total 30 minutes. And in the last evaluation, we asked the patient to move as much as possible within the mask, and then we took an X-ray. And the differences between at each checkpoint and the reference position was considered to be the movement.
And because we use the ExacTrac X-ray, so the question is how accurate is this system, in detecting the skull position? And there were many studies, but we also did them to confirm this issue. And we used the 3D phantom driving system, we are originally developed, and this system is basically developed to see the impact of the movement onto the IMRT dose distribution, and also to validate the ability of the tracking validation of the machine, it's a new machine we are now developing, but this system is also very useful in checking the accuracy of the ExacTrac X-ray because this is mechanically very accurate in positioning, it's about plus-minus 0.02 millimeter in static condition, even if we load the heavy phantom.
So, we set the scull phantom on this driving system, and we moved these phantoms to different directions and different amount, and at each point, we took an X-ray by the ExacTrac X. And differences between the known distance driven by the system, and the corresponding outputted value from the ExacTrac X-ray was considered as a positional error. And this is a result of the study, and if we used the CT of slice thickness of 1.25-millimeter, the average translational errors were around 0.1 millimeter with a max value of 0.5.
And this driving system cannot make any rotational movement in any directions. So, these values should be 0, but actually, ExacTrac X gave us some values. So, we counted to these values as also the rotational error, that was around 0.1 to 0.14 millimeter in average. So, we concluded that the ExacTrac had basically these uncertainties in detecting the position.
And in addition to that, if we make fusion several times, then the ExacTrac gave us different answers, and the difference were around 0.1 to 0.2 millimeter. So, combined with the results of the previous study, we think that the potential of the uncertainties in the ExacTrac X-ray system in detecting the skull position is around 0.2 millimeter. And in this study, we made five fusions and calculated the average position, and used the position as a checking point.
And this is the outcome, and this indicates the systematic errors of 10 patients, and the mean value was about around 10.1 millimeter or degrees. And the SD value was also very small, which is considered to be larger segment of our patient group.
And this is our random errors. And again the mean value was around 0.1 millimeter or degrees, which represents small segment over our patient group. And in here, we simply plotted the displacements of all checking point of the 10 patients.
So except one case points, all the translational displacements within 0.5 millimeter, and in 3D, the displacements were 0.09 in average, and with a SD value of 0.22 millimeter. So they were very small. And this is the rotational outcome plotting. And actually, in one patient there were larger displacements. So these are from that particular patient. So except that patient, the rotation errors were within almost 0.4 degrees.
And this is an example of a patient fixed in the BrainLAB mask system, and this indicates the time course of the motion of each components, and this patient is very stable. All the movement is within 0.2-millimeter degrees, and more than half of the patients with the BrainLAB system is like this. And this is another patient, and this patient had a relatively larger movement, but still almost within the 0.5 millimeter or degrees.
And then we made a comparison with the regular head and neck masks, we are usually using for the conventional fractionation treatment of the brain. And if possible, we attach the bite block here. And in this study, among 10 patients we checked, the 8th patient had a bite block fixation. And this is a result of the comparison about the systematic errors. And as you can see here, the craniocaudal direction with the regular mask is significantly larger, it's about 2.3 millimeters in average. So this is not acceptable for the stereotactic radiation.
And also this is a random error, again the random error in the craniocaudal direction is significantly larger, with a regular mask system.
And this is an example of the movement fixed with regular masks. It's this patient with a bite block. And the movement of this particular patient is relatively small. And in this case, after asking the movement, the vertical errors becomes a little bit larger. And this is another patient who had a bite block, and the movement is much larger.
And to visually confirm the outcome, we made a video continuously dissipating the X-ray images, and as you can see here the movement of the patient mouth, but the skull is very stable. And this is another patient fixed with the BrainLAB system. And there was a small movements of the patients' skull. This patient had a relatively larger movement, and this is a patient fixed with regular mask with bite block, and you can see the larger movement compared to the BrainLAB system.
So in conclusion, the intra-fractional movement of the skull within the BrainLAB head and neck mask system for the 6D Robotics were enough small for frameless stereotactic irradiation. Thank you very much for your kind attention.
And for frameless stereotactic irradiation, of course, it is very important of the reproducibility of the skull portion. And in addition to that, it is also very important, the intra-fractional movement of the skull should be small. And this high reproducibility was achieved by the ExacTrac X-ray system. And we also validated the overall accuracy of that one. But today, I'm going to concentrate on this intra-fractional movement. So, the purpose of this study was to evaluate the intra-fractional skull movement, fixed in the BrainLAB head and neck mask system. And we used the Novalis and also the ExacTrac X-rays 6D and the BrainLAB head and neck mask system. And we used a CT slice thickness of 1.25 millimeters, with a field of view size 40 by 40 centimeters, and the CT scan was taken with G right speed. And we evaluated the 10 patients for SRS or SRT with this mask system. And what we did was monitor the skull movement with the ExacTrac X-ray. And at first, we localized and corrected the position, by the ExacTrac X-ray. And after that, we verified the positions every 3 minutes, in total 30 minutes. And in the last evaluation, we asked the patient to move as much as possible within the mask, and then we took an X-ray. And the differences between at each checkpoint and the reference position was considered to be the movement.
And because we use the ExacTrac X-ray, so the question is how accurate is this system, in detecting the skull position? And there were many studies, but we also did them to confirm this issue. And we used the 3D phantom driving system, we are originally developed, and this system is basically developed to see the impact of the movement onto the IMRT dose distribution, and also to validate the ability of the tracking validation of the machine, it's a new machine we are now developing, but this system is also very useful in checking the accuracy of the ExacTrac X-ray because this is mechanically very accurate in positioning, it's about plus-minus 0.02 millimeter in static condition, even if we load the heavy phantom.
So, we set the scull phantom on this driving system, and we moved these phantoms to different directions and different amount, and at each point, we took an X-ray by the ExacTrac X. And differences between the known distance driven by the system, and the corresponding outputted value from the ExacTrac X-ray was considered as a positional error. And this is a result of the study, and if we used the CT of slice thickness of 1.25-millimeter, the average translational errors were around 0.1 millimeter with a max value of 0.5.
And this driving system cannot make any rotational movement in any directions. So, these values should be 0, but actually, ExacTrac X gave us some values. So, we counted to these values as also the rotational error, that was around 0.1 to 0.14 millimeter in average. So, we concluded that the ExacTrac had basically these uncertainties in detecting the position.
And in addition to that, if we make fusion several times, then the ExacTrac gave us different answers, and the difference were around 0.1 to 0.2 millimeter. So, combined with the results of the previous study, we think that the potential of the uncertainties in the ExacTrac X-ray system in detecting the skull position is around 0.2 millimeter. And in this study, we made five fusions and calculated the average position, and used the position as a checking point.
And this is the outcome, and this indicates the systematic errors of 10 patients, and the mean value was about around 10.1 millimeter or degrees. And the SD value was also very small, which is considered to be larger segment of our patient group.
And this is our random errors. And again the mean value was around 0.1 millimeter or degrees, which represents small segment over our patient group. And in here, we simply plotted the displacements of all checking point of the 10 patients.
So except one case points, all the translational displacements within 0.5 millimeter, and in 3D, the displacements were 0.09 in average, and with a SD value of 0.22 millimeter. So they were very small. And this is the rotational outcome plotting. And actually, in one patient there were larger displacements. So these are from that particular patient. So except that patient, the rotation errors were within almost 0.4 degrees.
And this is an example of a patient fixed in the BrainLAB mask system, and this indicates the time course of the motion of each components, and this patient is very stable. All the movement is within 0.2-millimeter degrees, and more than half of the patients with the BrainLAB system is like this. And this is another patient, and this patient had a relatively larger movement, but still almost within the 0.5 millimeter or degrees.
And then we made a comparison with the regular head and neck masks, we are usually using for the conventional fractionation treatment of the brain. And if possible, we attach the bite block here. And in this study, among 10 patients we checked, the 8th patient had a bite block fixation. And this is a result of the comparison about the systematic errors. And as you can see here, the craniocaudal direction with the regular mask is significantly larger, it's about 2.3 millimeters in average. So this is not acceptable for the stereotactic radiation.
And also this is a random error, again the random error in the craniocaudal direction is significantly larger, with a regular mask system.
And this is an example of the movement fixed with regular masks. It's this patient with a bite block. And the movement of this particular patient is relatively small. And in this case, after asking the movement, the vertical errors becomes a little bit larger. And this is another patient who had a bite block, and the movement is much larger.
And to visually confirm the outcome, we made a video continuously dissipating the X-ray images, and as you can see here the movement of the patient mouth, but the skull is very stable. And this is another patient fixed with the BrainLAB system. And there was a small movements of the patients' skull. This patient had a relatively larger movement, and this is a patient fixed with regular mask with bite block, and you can see the larger movement compared to the BrainLAB system.
So in conclusion, the intra-fractional movement of the skull within the BrainLAB head and neck mask system for the 6D Robotics were enough small for frameless stereotactic irradiation. Thank you very much for your kind attention.