Optimizing the Introduction of AlignRT- Trouble Shooting

All right. Optimizing the introduction of AlignRT. This is a troubleshooting. So about us, we’re from California, Sacramento area. We have four sites: Auburn, Roseville, Sacramento, and Cameron Park. We’re all Varian TrueBeam, AlignRT. We have RayStation treatment planning system now. We are tattoo-less and mark-less head to toe for all patients. Kind of way down the line for new users, but now we also have a scripting process within RayStation, so our physicists actually provide us with isocentric coordinates. So we’re able to go right to isocenter on day one and use AlignRT to put our patient exactly where they’re supposed to be.

With AlignRT playing such a vital role in our clinical workflows across all of our locations, we recognize the need for better consistency and collaboration between all of our locations. To help optimize its use, we created the AlignRT Champion Group. In this group, it’s made up of therapists within our department, and the goal is to help standardize practices, encourage open feedback, and support ongoing troubleshooting and workflow improvements. This has really been the foundation for us in moving from simply using AlignRT to optimizing it across our entire system.

Today, Justine and I will walk you through some of the common clinical scenarios we focused on for troubleshooting patient positioning with AlignRT. These are areas where we saw variability across sites or opportunities to improve consistency and efficiency in our workflows. We’ll be covering how we’re using regions of interest, how to approach treatment reference capture, and how we’re utilizing the center couch function in a few different clinical situations, such as free breathing, breath hold, and even treating multiple isocenters from a single cone beam. Then we’ll finish with a troubleshooting tip for our head and neck cases using postural video.

So starting with regions of interest, this is an area where practices varied. However, a well-drawn region of interest, or ROI, really has a significant impact on how the performance of AlignRT works in our clinics. When thinking about what makes a good ROI, you want to make sure that at least two cameras are consistently visible, or the ROI is consistently visible to at least two cameras at a time. This ensures that you’ll have reliable tracking. When drawing the ROI, it’s important to cover a relevant area, ensuring that the ROI isn’t too large where it will introduce noise, and not too small where the ROI will lose stability.

We found it’s helpful to include clear, prominent surface features while avoiding anything that may throw off tracking, like irregularities or anatomy that may change over the course of treatment. Another big piece is being aware of what the ROI can actually see. So you can check your postural alignment to confirm if parts of the ROI are being blocked, possibly by your gantry, and edit those areas out. Overall, there really isn’t a one-size-fits-all. The ROI should be tailored to the patient and the setup. Therefore, therapists are encouraged to feel comfortable adjusting the ROI on the fly. Anytime you do make a change to the ROI, it’s important to re-image and confirm the new ROI is tracking accurately before treating.

Building on adjusting regions of interest to optimize tracking, here are a few practical ways we adjust the regions of interest. In some cases, to ensure that the ROI can be seen by two cameras, we adjust the ROI to the contralateral side of the patient, which are located here as examples. And in other cases, we adjust the ROI more mid-lateral. Both these examples allow better visibility from the side camera. And as mentioned in the previous slide, we want to focus on excluding surfaces that may change over the course of treatment. In this example, we have drawn an inverted T to include stable surface area for tracking. We also focus on including prominent surface features, such as this example, where we’ve included the gluteal cleft in the ROI.

Once we have an effective ROI set, we verify the external contours with our postural video. You can see here in these images, the contours are displayed by the magenta lines. Using the postural video helps make sure what we’re seeing on the external contours matches what the ROI is actually tracking. If the ROI is not tracking appropriately, then the anatomy and the magenta lines would not be matching. If this does take place, the option is to adjust our ROI on the fly to improve the tracking, whether that’s reshaping the region of interest or editing out areas that don’t contribute to the stable tracking. And anytime, like I mentioned before, anytime you do make a change, we make sure we re-image and confirm that the new ROI is tracking accurately before treating. All these adjustments are about improving accuracy and consistency to optimize tracking the patient’s position with AlignRT.

Moving on to treatment reference capture. This is another way we have optimized our use with AlignRT for troubleshooting patient positioning. I’ll go over the function of treatment reference capture and how we’re using it. Treatment reference capture gives us a way to reassess the patient’s position in real time and work from a more accurate representation of where they are that day. It’s essentially a 3D still frame of the patient showing the patient’s current position compared to their simulation position. Once you snap a treatment reference capture, you’ll see the planned position in purple and the patient’s current position displayed in green. So this is helpful because it provides a snapshot of any positional differences, which allows you to quickly identify what adjustments need to be made.

While you’re setting up the patient, you can snap a treatment reference capture multiple times. So with this, you can reassess the patient’s position as you make changes. Unlike an SGRT capture, this isn’t saved, so it’s essentially just a real-time tool to help guide positioning in the moment. So I’ll walk you through the steps. We’ll start by opening the patient in treatment and monitoring as usual. From there, we’ll align to isocenter using our standard workflow. Once we’re comfortable with the patient’s alignment, we’ll pause monitoring and capture a treatment reference. You do so by clicking the icon with the camera on it. Then we’ll resume monitoring and continue with setting up the patient using that new reference capture to guide any positional adjustments, keeping in mind that the purple is the planned position.

Looking at this patient here, this is a good case where treatment reference capture can be helpful. As you can see in the shoulder and arm region, the patient’s current position isn’t matching well with their planned position, and this is something that we see often with arms and other extremities. So we found this tool especially helpful in cases like these where anatomy positioning can vary from day to day. All right, we’re going to go over the Send to Couch feature next. What is Send to Couch? Send to Couch essentially acts like imaging, like a CBCT or a portal image. You can send shifts to the couch and adjust your patient in 3DOF or 6DOF. A lot of our presentation today is tailored to having a Varian with 6DOF. However, you can use it with 3DOF also.

So, these are some tips for when you can utilize Send to Couch. So, we all have those patients where every day they have that same little bit of roll that you can’t quite achieve manually. You could use Send to Couch in that manner. You walk out of the room, your patient has sneezed or coughed and shifted a little bit. You can Send to Couch and get them back into tolerance before you treat or before you image. We have the patients settle, of course, when we’re all in the room and adjusting them. They’re pretty rigid. And then you walk out, and all of a sudden they’ve relaxed into place. You can Send to Couch and get them back in tolerance. And then, kind of a unique situation, but you have mono isocentric treatment with more than one prescription. You can use this also if you have to close out of one prescription and go back into another and you’ve already imaged. You can use your Send to Couch to stay at where you know you’re in the right place.

So it acts like an imaging shift. Like I said earlier, it’s comparable to a cone beam or portal image. You can use that to physically override your couch if you need to make adjustments. For us, it’s based off of the Varian tolerances and the tolerances that your physicists put in place. It can influence what type of shifts you’re able to make, so typically we’re within three degrees and two centimeters to be able to Send to Couch, which really you want to be within your deltas anyways.

So this is a step-by-step for how you use the Send to Couch feature. So if you’re using a Varian, again, very tailored to Varian, what we typically do as we check in our patients and do our timeout every day is we check them in, and then we go ahead and hit Prepare while we’re at the console. So once we go into the room, we’re able to use the Send to Couch feature if we need to inside the room. You align your patient. You want to get them within the deltas. But if you are having trouble with that, this is where you can use that Send to Couch. And we’ve gotten in the habit now, if we could be better, why not be better? So even if we’re within the deltas, we try to just go ahead and send 6DOF and get everything as close to zero as possible. You Send to Couch, and it overrides just as imaging would, puts the patient right in tolerance.

So we’re going to go through a couple of ways you can utilize Send to Couch. Like I said, we try to just utilize it in general. Get the patient as close as you can, go ahead and send 6DOF. If we can be at zero, why would we not want to be at zero? But this one is for two Isos from one cone beam. So this is our example here, so you can see we have one Iso here and one Iso here. We can see both in the same cone beam. So what we did in this case is we used the Send to Couch feature so that we didn’t have to image the patient twice, since we could see it in the one cone beam. We cone beamed the first Iso in the first script and made sure everything was on, and then when we go into the second prescription, we applied manually the shifts from the plan, and then we used that Send to Couch feature to send a zero essentially to couch, telling the couch, “We don’t want you to move. We know you’re in the right spot.” And then we were able to treat the second Iso as well without having to give that patient additional imaging.

Troubleshooting Send to Couch for breath hold. A big question that we get is can we send the vert on breath hold? And you never want to send your vert on breath hold, because we always want to make sure we have that same displacement as we did in simulation. So you should only ever send your vert on free breathing, and that ensures that your patient is taking the same breath that they took in simulation. So one of the nice features with Send to Couch again is patients are taking this breath, they’re getting tired, they’re shifting a little bit. You can use the Send to Couch function from outside of the room. So if you notice someone settling the longer that they’re on the table, you can go back to your free breathing. One thing to note is you always want to put your free breathing in as a setup only. That will actually make it to where you’re not able to treat your patient accidentally on your free breathing ROI. So you can go back to your free breathing, see if they’ve moved at all, see if they’ve settled in. You can Send to Couch again if you’ve noticed that your patient has settled a bit, and make sure they’re achieving that breath hold.

For our final example of how we’re optimizing the use of AlignRT for patient positioning, we’ll look at a technique we apply to our head and neck cases. We focus on aligning the patient without the mask first, which gives us a more accurate baseline of the patient’s position. You can use treatment reference capture at this point, but we rely heavily on the postural video when setting up our head and neck cases. We set up the patient first without the mask, and this helps guide any positional adjustments. This allows us to reposition the patient’s chin to correct for pitch and/or adjust their shoulders to correct for roll and rotation. These areas are where we tend to see the most variation in setups. And these small adjustments that we’re making prior to placing the mask on the patient help improve consistency and tracking accuracy during imaging and treatment, especially for clinics that don’t have a six DOF table, where you are unable to correct for pitch and roll during imaging. And if this gentleman here looks familiar, he’s in the front row right here. He’s our coworker, Josh. And you get a signature.

And we hope this presentation has provided some practical ways to help optimize the use of AlignRT in your own clinics, especially when troubleshooting patient setups. And for those of you that don’t have AlignRT yet, we hope that this presentation demonstrated how AlignRT can be beneficial when it is implemented into your site workflows. So thank you for your time and attention.