Advances in AlignRT: Enhancing SGRT Precision, Safety, and Workflow

Adi Robinson, Senior Medical Physicist (00:04):

These are my disclosures and what we do you’ve already seen what AdventHealth looks like just as a different map that Mike uses. This is where Mike lives. This is where I live, you know, Mickey runs the state. So but important note about AdventHealth is that we really focus on innovation, safety and the patient experience. We put that as our number one goal. And part of what I’m going to talk about is carving that path using AlignRT. First of all, you can have biometric facial recognition to eliminate your timeout if you purchase the patient AlignRT to really, really emphasize these steps. So, for those who don’t know what SGRT is, we register a time 3D surface of a patient to a reference surface comes from CT. We use camera projections to generate that surface in real time, and then compare that to the patient’s positioning on our reference dataset.

Adi Robinson, Senior Medical Physicist (01:06):

By doing so very, very quickly, we still get sub-millimetre accuracy as Mike detailed in his commissioning talk. And, and that’s, it’s pretty amazing. So with the out of the box AlignRT you get the following: You get enhanced patient set up add-on where you can just capture the patient’s face on day one and that kind of interlocks and gives you that no need to ask for their birth date and name and all that social security stuff every day before you treat them. You get Postural Video, which is this, this view where it’s kind of an outline of the patient. So you can set up the patient based on your six degrees of motions, your real-time deltas, which are based on the ROI that you drew.

Adi Robinson, Senior Medical Physicist (02:09):

But you also have this full body view to tweak the positioning of the arms, the legs, the knees, the abdomen, the arch of the back, whatever you need to do in order to kind of perfect the setup. And by doing so, you basically eliminate tattoos. You’ll hear about today and tomorrow about all these facilities that are tattoo and markerless. They don’t use that anymore in their facilities because they set up to the surface, not those three random dots you put on the patient. Adaptive monitoring deformation view, which you’ll see here, kind of compares how the surface looks today versus how the surface looked at CT. And it gives you some sort of metric, so you can see high and low spots. You can see if the patient has swelling or then doesn’t have swelling anymore, if they gained weight, lost weight, anything in that surface changed.

Adi Robinson, Senior Medical Physicist (03:00):

It also helps you optimize your ROI. So if you see a lot of red or blue areas in your ROI and the patient’s supposed to be DIBH or free breathing or whatnot, you can kind of customize your ROI for best utilization for that patient. So I think it’s a very underutilized tool, but it’s, it’s, it’s very, very, very powerful. And you have motion management. So you can either use the respiratory gating module right here if you want to do amplitude or face-based treatments. But if you’re just doing DIBH, you can use the regular-time motion with beam hold, and you can do DIBH. So all of that is AlignRT out the box. Probably some of you are familiar with it but it also comes with some challenges. It does still provide precise setup and monitoring.

Adi Robinson, Senior Medical Physicist (03:52):

However, as we just saw, the ROI can be challenging. It is manually created and is very, very subjective. Some therapists like it one way, some therapists like it the other way, there’s no kind of uniformity across the board. Some have more experience, some have less, some spent 10 minutes drawing it, some spin, three seconds drawing it. So there’s no consistency and that leads to trouble down the road when the patient actually shows up for treatment documentation. So if you need to prove that you actually treated that patient, that fraction with AlignRT and you can have continuous, you have a manual process of doing so. So you have to generate the report at the end of the fraction, import it outside of the AlignRT system and into your record and verify system while the next patient waits for you to finish doing that. So it is a kind of clunky workflow, and if you have to document and if you have to do it for billing or for audits or whatnot, it takes time and it’s not so simple.

Adi Robinson, Senior Medical Physicist (04:55):

And then you have limited plan or dose visualization during the setup of the patient. So before you even turn the beam on, you kind of want to make sure that you’re setting them up correctly. And an overlay of what the plan is going to look like would be very, very helpful. So, lo and behold there is some AlignRT innovations that address those workflow limitations. So with Beam Guide, we have a visual overlay of the patient’s plan over Postural Video with the OIS reports, we have an automatic documentation module, and then Auto ROI gives us a fast and accurate auto ROI generation. So let’s start in the beginning when we’re going to, you know, we just prepared the patient and we’re going to draw the ROI, you know, normally the ROI for those who don’t know the region of interest is that area on the body of the patient’s body that the system’s going to compare current versus reference positioning.

Adi Robinson, Senior Medical Physicist (05:51):

So that’s where the three six degrees of real time deltas come from. And as emphasized earlier, and you’re going to hear that a lot today, a really well-defined ROI ensures that AlignRT can accurately track the patient’s positioning. If you have a garbage ROI, it’s going to be really, really bad RTDs and really, really bad setup. So you’re setting the tone here for the patient’s treatment. So with auto ROI per protocol, it will give you about three to five options of what the ROI should look like, and you can preview those on your actual patient and then decide which one works best for you. We’ve, throughout our testings, that about 90% of ROIs are clinically acceptable with almost no changes to them. But the module does give you the option to edit it. So you can, if you see something that you like but you want to tweak it a little bit, you can still edit to it.

Adi Robinson, Senior Medical Physicist (06:48):

If you want to start over, it can erase the whole thing and kind of try again. It does introduce consistency and reduces creation time. So let’s look at a few examples. So we treat a lot of breast patients in our clinic. So here’s an example of our right and left breast. You have the autogenerated ROI here, this is the manually drawn ROI that we use clinically. This is the overly difference. And you can see it’s about 90% similar ROI. Now what we did is we set up the patient in a few, a few fractions using the auto ROI made sure that all the deltas are happening right before treatment. We flipped to the manual just to see how the deltas are changing, if we can just continue treating. And we saw no difference with, with breast plan.

Adi Robinson, Senior Medical Physicist (07:35):

So that gave us the confidence that the auto ROI is is is very, very useful. With brains and an open face mask, it’s a little bit more challenging because the ROI, depending on how you make your mask, you know, how lateral do you go and, and all of that. But with all that said, again, looking at two cases here we are a lot more generous in how we we do, at least from this view, but it’s still about, you know, 85% similar. And again, it give us good clinical results. So, you know, these brain ones might need a little bit more tweaking, but it’s still a five second generation time versus, I don’t know, a minute or two. So I don’t know, count that counts with something. Pelvises are a little bit more complex because depending on your clinical needs some, in some cases patients, we use this band over the pelvis and some of them we don’t.

Adi Robinson, Senior Medical Physicist (08:33):

Some of them we just focus on the thighs. So that’s kind of a selection that you have to make clinically, but both Auto ROI generation generated ROI’s have either just thighs only or thighs with band. So you kinda have to decide what works for you. And this shown good success for us. So kind of to summarize that we are able to implement it across broad treatment sites. So it currently, you have auto ROI protocols for breast DIBH free breathing chest, abdomen, and pelvis. They’re all headfirst, supine but I’m guessing more and more ROIs will come as the as more implementation of this module will happen. And, but what we’re really excited about is this consistency, consistency in standardization. Now we have a minimally good starting block for ROI creation.

Adi Robinson, Senior Medical Physicist (09:37):

So I can bring a therapist straight of out school, he’s never seen SGRT in their life and I can explain this better to them than the old flip book that sometimes can be confusing. So we have a minimum standard that we like and then we can just adjust it from there. We don’t care that much about time savings. We have an offline, you know, workstation that we can use not during treatment, but it’s the quality of the ROIs that we want to elevate. And now we’re start, our starting point is much higher than what it was before. So that is kind of what we wanted for, we’re looking for. So beam guide once the ROI is generated, we’re ready for treatment of the patient. So we need to set them up. And what now the add-ons to Postural Video is twofold.

Adi Robinson, Senior Medical Physicist (10:21):

We have a, a plan outline which just shows the outline of the patient here in white on top of the postural video that you see it in, in purple. And we also have the, the planned projection. So this has got an active projection of where the, the beam is going to go. So kind of imagine of where the dose is going to be deposited, kind of thought. And those, they can exist together or you can turn one of them off, whatever is confusing. But now you can set up the patient based on where the beam is going to go. So, being a physicist, we like to validate things. So before we start using it, I kind of generated plans with certain shapes and I wanted to see how well can beam guide can reproduce these shapes. You know, I made classic shapes, but then I started thinking, we need to make a mistake.

Adi Robinson, Senior Medical Physicist (11:13):

Will it catch a planning mistake? So I created a circle, a bad circle with a leaf pulled out and a leaf pulled in just to see how it goes. We have the luxury in my clinic that we have dose RT as well. So dose rt as you’ll see today gives you kind of a, a verification of what the dose, what the, what the dose you delivered. Kind of it’s an end of treatment dose visualization during the end. So I can actually use Beam Guide to validate DoseRT and DoseRT to validate Beam Guide. So it’s kind of a, a cross-contamination thing. So here is, oh, video’s working good. So here’s me looking at the field and beam guide postural video view. So I’m kind of loading up the triangle and the diamond and there’s the L shape

Adi Robinson, Senior Medical Physicist (12:09):

And the circle which all look good. And then I want to see what the bad circle look like because that’s the important part. And it does pick up that there is an MLC leaf out of the way in and out of the way out. And I can turn off the projection if I wanted to just to kind of focus on the, on the field light, but this is what I wanted to see. I didn’t want see something that kind of ignores that. So that’s good. So if the therapists are setting up the patient and they see something that is unexpected before they even beam on, before they did anything, that is kind of a hard stop for us. And then they can come to physics and say, Hey, I expected the plan overlay to kind of be on top of the breast, but I’m seeing it on her chin or something.

Adi Robinson, Senior Medical Physicist (12:50):

We need to kind of look at it, and then we can figure out if it’s a setup issue or a plan issue. But we haven’t harmed a patient yet. We haven’t delivered anything. So that is a very important tool. So here’s a bit of a DoseRT versus Beam Guide kind of comp comparison where this is the, I delivered all these plans and this is what it looks like. This is a cumulative fraction image. So you can see the triangle correlate to the triangle. The diamond correlate to the diamond, the LS shape correlates to the L shape. And this is the two circles, the good circle and the bad circle. And the error is visible in both, which is good. That means you could see it either before you set up the patient or after you set up the patient, but you want to see it before you set up the patient especially if it’s a planning error.

Adi Robinson, Senior Medical Physicist (13:42):

So here’s some real time, real live cases that we’ve used it for. You can see this is a VMAT case where we just treated an arc to arc VMAT. I’m expected to see a band across the patient’s pelvis, and that’s what I’m seeing. This is a prone breast case. I know you guys don’t treat pros a lot, but this is kind of one of more interesting case that I’ve found because we set up this patient Postural Video, looks awesome, rt, the six degrees, deltas look awesome, everything’s in the green, everything’s good. But this little sliver of breast is not covered by Beam G

uide, which means it’s going to get underdosed. So we had to fix her rotation. There was something, there was an issue there. And once we tilted her a little bit, her entire breast was in view and then we delivered the fraction correctly.

Adi Robinson, Senior Medical Physicist (14:38):

So this is something that Postural Video, none of the features of AlignRT right now was able to catch, but beam guide and that raised our confidence in the system, you know, through the roof. Here’s a supine DIBH breast, and again you can see in three different views how the breast looks like and then how it was delivered. And the shapes look the same. And that’s kind of what I’m looking for here. That the pretreatment shape looks the same as the post-treatment shape looks the same as in the treatment plan, and that’s ideal for me. Here is kind of a challenge that we like to do. Sometimes we get stuff to test and I do a blind challenge for the therapist where I get a piece of equipment and I kind of tell them to test it and I want them to the real reaction to them.

Adi Robinson, Senior Medical Physicist (15:26):

So we got this clear bolus to test and I figured this super clave and right breast, left breast patient would be the best candidate for it. Now we do it mono iso ally, but we don’t bolus the super clave. And I kind of wanted to see if they’re able to use Beam Guide to set up that bolus. So in Beam Guide, you can select which field you want project. So you can do all fields, you can do just the super clave, or you can do just the, the breast. So the whole idea was put, turn on the super clave, only put in the bolus, make sure the super clave is not bolus, then turn off the, the, the left breast, make sure that it’s all in the bolus, then turn all of them, see how it looks and then treat. And this is the resulting treatment, but it doesn’t really tell us much.

Adi Robinson, Senior Medical Physicist (16:11):

This is pre-bolus setup, this is post-bolus setup. You can see how clear and shiny the bolus is. So if we look at the actual delivery of the beam, this is the Taren cough signal coming from DoseRT. And you can see we’re treating the super clave. It’s hard to see, but this is the outline of the bolus. So they set up the bolus perfectly without seeing that bolus before. We saw almost not no beam guide experience at all. And they did right off the bat. Perfect. So that’s how easy it is to use Beam Guide in setup. And I was very impressed that they were able to do that. So kind of talking about Beam Guide in general it’s easy to inter if you use Postural Video, it’s easy to integrate in all of your workflows right away.

Adi Robinson, Senior Medical Physicist (16:57):

And it’s super, super, super helpful if you have any accessories or kind of a budding fields and you got to have to worry about a gap or an overlap or something like that. It gives you immediate visual confirmation of your beam entry and exit, which is super, super, super useful. And you know, our, I quoting my therapist here, they said it’s a must have. And they love seeing the plan at set up. So it’s, it’s, it’s a very, very valuable tool. So now once we treated the patient, we need to document that we treated the patient. And this is where the OIS reporting module comes into play. We have Aria, but it also works on Mosaic. So, but this is an ARIA snapshot. This is the treatment basically with CBCT, the four arcs. And then now the align RT reports are going to pop in at the end automatically.

Adi Robinson, Senior Medical Physicist (17:48):

You don’t have to export, import anything. It just set up for once and it’s going to do it forever. If you click on the screenshot, it shows you the, the screenshot that it takes at the last second of treatment. So you can prove that everything was within tolerance. And if you want to see the full report, it’s right here. It has the patient’s name, MRN, how long was he monitored for, you know, you can have the delta graphs of how much he deviated and all that stuff. And all of this is available for the physician to approve as they’re approving their IGRT images. If you’ve taken them that day, they can approve the SGRT images all at the same time. So that’s really, really, if you bill for it or if you need any kind of documentation for audits, you can see that the physician reviewed both IGRT and SGRT at the same time before the next fraction it was approved.

Adi Robinson, Senior Medical Physicist (18:39):

And here’s the documentation that we actually did that. So, all in all, about the new modules, I think the integration of the new modules represents a significant advancement in AlignRT. Though the tools in really, really enhance treatment verification, documentation and automation and really, really improves the position safety and efficiency of our workflows. And I think I would recommend you guys using it because it really, really makes the clinic life a lot more efficient and simpler and just, just always going to be something more. So it really, really makes the complex a lot easier. So with that, thank you to Mike. And yeah, this is me on a normal day. So thank you.