Investigation of Patient Positioning Times and Remaining Positioning Deviations Using AlignRT InBore on a Varian Halcyon Accelerator

Niels Goetting (00:04):

We had once in a life it took about four to five months to get the SGRT workflow running, mostly due to management reasons at our side it took about four to five months to get the SGRT workflow running due to, mostly due to management reasons at our side, not Vision RT. So we could take this chance to take data without SGRT and later on with SGRT. In the meantime, we got the, the Markerless Award from Vision RT. So we are now completely markerless.

Niels Goetting (00:43):

We are located in the northern side of Germany, and our site is the third one of our group of radiation therapies in Flensburg. So at the very top of Germany, it’s still a very small site. We have only one accelerator. It is a very unhealthy machine and a CT. So we have both AlignRT and SimRT running in our department. It’s a small team of two physicians and physicists, and RTTs. So very small but very enthusiastic. So we can do lots of things.

Niels Goetting (01:27):

And our way with SGRT is rather short, but really nice. We started in March of last year, so one and a half years ago. Our centre is located in a shopping mall. This, which is kind of unknown, but it’s really nice, especially for patients and taxi drivers, as you can imagine. Also, for us, of course, we get all things. And in August, just a few months later, we got a AlignRT and SimRT, so we started with SGRT and May this year marks the markerless award. So, markerless radiotherapy. The variant healthy machine is a little bit strange, as several of you may know it’s in a ball-type linac where you have some kind of special philosophy. You put the patient on a set position outside the machine so you can easily work on the patient. And then afterwards, you drive the table inside the machine, so you have a treatment position. And therefore we have four AlignRT this nice in boring room with another couple of cameras and projectors. And so we have more or less two systems, which is of course, special.

Niels Goetting (02:50):

Another special thing is that we only have a 3D couch, so we have only translational movements possible. So it’s the first group in the AlignRT screen, which we can use to correct the position of the patient. And therefore our main topic to work on is of course to reduce rotational deviations seen by AlignRT of course also seen in the CBCT afterwards, but you usually don’t want to repeat the CBCT only in case that is really not acceptable. So AlignRT is the instrumental work on these rotations.

Niels Goetting (03:33):

When we thought about what to do with this data set, or the time when we did not have SGRT on this new machine, we thought that the patient positioning times could also be interesting, of course, as SGRT is for positioning of the patient, but the time to use for this positioning could also be interesting. So we took some stopwatches. In fact, we used a digital one, not a very old-fashioned one, but of course, it’s okay. And then we defined a time interval to measure. So when the patient lies down on the couch, it’s on the down position. And then in the Varian bore, you have to follow the blue buttons. So it’s, there’s a setup button. The table moves up to a set up position and when pressing this table one of the RTTs press the stopwatch to start it. And after positioning, you have to press a load button to move in the table into the machine. So that’s our standardized stopping point. So we have a protocol for measurements, which is really clear. And so we have comparable data. We don’t want to start earlier because the way of the patient enters the room, that depends on the patient. Okay, so he’s slow or he’s running to the table. It’s not interesting for positioning. And, the stopping point later than reaching the treatment or driving to the treatment position is also not interesting because we don’t want to measure the time for the CBCT or the time people need to interpret the CBCT. The measurement time itself is written down on paper outside the bunker. So here’s an example of a protocol. We take it from our area database, from the time planner, from the patient’s times when they should come, but it’s anonymized. We just write down the date, the time of the day, and the region of the body so we can investigate diff different positions on the body. Then written down by hand. And also, some remarks could have been that something happened, which leads to not using this special measurement. Example for the remarks of the renewal of skin marks we used without SGRT, the plasters had to be renewed, as well as difficult or nervous patients follow those materials. All these things were written down, and you may say, okay, this is old-fashioned to write it down on paper or RTTs, were also nice to take these measurements. And they told me that’s the easiest way for them. So we got it on a sheet of paper each day, and then we just typed it down into some kind of spreadsheet. So it’s a very feasible method.

Niels Goetting (06:49):

And now for the data set we without SGRT, we had a number of more than 700 fractions with SGRT, and about 400 fractions, just a few had to be thrown out due to remarks. So it’s a nice database for analysis. This first set with SGRT will be continued in the beginning of the new year due to some organizational points.

Niels Goetting (07:20):

So in total for the analysis, we have split it here down to chest, which is mainly breast pelvis, which is mainly prostate and some extremities like knees and feet. And in total, we have nearly 1,100 single measurements, single fractions where they went into the room, press stop, a start and stop, wrote down the time of the measurements. And I think it’s time to say thank you to the four RTTs who ran around and took this data.

Niels Goetting (08:00):

So now for the data itself, I show here the positioning time over the region of the body, chest, pelvis, and extremities in the beginning of the life of the side there were not so many abdominal or head patients. So you see a violin plot in blue. The left side of each distribution is without SGRT, the red one is with SGRT. So you’ll see here clearly trends that maybe the time for positioning has become shorter. So promising trends have a look on these histograms for the single parts of the body for the chest. So breast you see in red now with AlignRT in blue before, and we see a time difference of about 28 seconds. So with SGRT we became really much, far faster. Same, more or less for the pelvis, for the prostate cases, mainly it’s also the same manner. It’s 29 seconds difference for the extremities; the significance is not that big, but it’s 23 seconds faster. So this is really promising and nice.

Niels Goetting (09:25):

Our management was happy to hear this, and we also looked on the distributions just here, repeat it in smaller values for these really longer times, mainly without AlignRT InBore. And we found out that it’s the skin marks that went away mainly during the weekends. So on Mondays we had to renew these marks and plasters. So this took of course, quite a lot of time. Also, some very heavy patients were difficult to move. And also some complicated and nervous patients who needed to be assured that everything is okay. Also, some patients needed bolus material, which of course, also takes some time. And yes, this is also faster now with AlignRT InBore. So I think it’s always worth to look on cases where you really take quite long. Maybe you find a reason which can be omitted so you become faster.

Niels Goetting (10:40):

Yes again, these numbers, so I can say it’s 25 to 30 seconds, which we are faster now, which is of course, great, especially for the patients. They don’t need to lie on the table so long, so half a minute, maybe short for one day, but for 20 or 30 fractions, it’s quite a couple of time of life. And the comfort is of course there. And also economically, it’s nice for us to spare half a minute.

Niels Goetting (11:11):

Then we also take the time to look at the position accuracy itself without and with SGRT on the Varian Halcyon accelerator; it’s mandatory to take an image each day. So MVMV or CBCT. So mainly we do CBCT. So we have a database for to play around with for investigations. Of course, we do an online match each day on the CBCT, but there are translation corrections only due to the 3D couch in the offline review, offline match. We can do it in the office after the fraction. We can also have a look on 60 corrections, which would have made everything even better. But we can, of course, not do it on our machine, but we can have a look on the rotation role and pitch, which are always, of course, given by AlignRT InBore, so we can investigate these necessary corrections.

Niels Goetting (12:24):

And also for this, we have a data set. We took or try to take 10 patients and all fractions of them. So we have some of about 600 fractions before and nearly 900 fractions now with SGRT. So it’s about 1,500 fractions. That’s quite a lot of data. My colleague Le Schultz was working on this. And we also made a comparison with our RTTs, who of course, work on this each day, each fraction. So we tested that all observers do the same work, and we made a protocol how to get the data out. And this is just a look on the ongoing analysis. And we see here, again, in blue the time without SGRT, without AlignRT and the red with AlignRT And we, especially for the left breast, we see some promising trends. So also as known in the community, you become better, of course, with AlignRT before the CBCT.

Niels Goetting (13:38):

So, to summarize this positioning accuracy, we save some time with the markerless workflow. Our patients have more comfort using no skin marks because they don’t see them each evening or whenever in the mirror. In the online workflow with the CBCT, we have fewer corrections now of course with AlignRT as expected, so we also save some time around the CBCT. With the RO rotations, we also see these trends. So we have a win-win situation for patients and also our department. The same for the positioning times. It was the main topic of this analysis we see about 25 to 30 seconds, and we are faster now. And this is, of course, an explicit advantage for patients and the department. So that’s really a win-win situation. And we are happy to have this AlignRT InBore. Yes, thank you very much.