Implementing Dose Visulization to Help Improve Patient Safety

Very glad to be here. And I will be talking on implementing dose visualization, which I really believe improves patient safety as well as plan quality in radiation therapy. First, my acknowledgments and contributors. I wouldn’t be here without the support of my physics team, who actively use this technology in many of our clinics. I don’t have any conflicts of interest. I am so passionate about DoseRT that my organization paid for me to be here today. Also wanting to recognize one of our medical physics residents, Ila Farhang, who contributed to this talk as well.

Advocate Healthcare has sites in North Carolina, Georgia, as well as in the Midwest, where I am primarily based, in Illinois and Wisconsin. We currently have five systems of DoseRT, two that are implemented in our Illinois market, two that are implemented in our Wisconsin market, as well as one in the atrium market. We are pending installation of three systems at our major cancer center in Milwaukee at Aurora St. Luke’s.

And really, I wanted to start by talking about our interest in DoseRT. We have been AlignRT customers since 2013. We have it at nearly every one of our system installations. And we were very excited when we heard about the technology, and even the concept of Cherenkov radiation at ASTRO. We feel as though it really is analogous to a seatbelt. You wear a seatbelt every time you drive a car. You really don’t notice it until you need it. I know I was involved in a pretty bad car accident about 30 years ago, and a seatbelt saved my life. And so I do think that as we look and evaluate DoseRT, that is really its benefit of showing you where the radiation dose is going.

It also aligns with our internal cancer service line strategy of really providing care to patients in a compassionate way to the loved one standard. And that’s very important to us. Many of us receive our own healthcare within our system. Many of us have friends and family that also receive healthcare within our system. And so we really always try to do the best possible job we can. And I think we do a really good job with patient treatment, but it can always be better, and good is the enemy of great. And so we are constantly looking to iteratively improve our processes.

The strong advantages of DoseRT qualitatively is the enhanced visibility and confidence that we gain in treatment delivery. What we see as the future, quantitatively, and Dr. Gladstone spoke a little bit about this yesterday, is the potential for measuring dose as the technology matures.

So our installation experience was very positive with VisionRT, but there are some elements that are outside or have been outside of VisionRT’s support. One of those is lighting. So if you are thinking about implementing DoseRT, you really need to think carefully about your lighting. So in August, we implemented DoseRT, and with one of our cameras, we were able to visualize dose. On the other camera, we were not. Through a lot of investigation, we found that that had to do with the UV light filtering on some of our lighting.

Additionally, every vault that I have seen is very different, and we have several different vault varieties. We have what I would say is a traditional vault. We also have a garden vault that is open to the outside. So I have a picture of it; it’s covered in snow. Not the greatest idea in an environment where you have all the seasons, although that natural ambient light is something we did for the patients, and we do have natural plants that we put in the springtime in that area.

So looking at our traditional vault, we have a lot of elements in this room. First, to highlight, we have two laser systems that are set for lateral positioning, really listening to our therapists wanting to have an ergonomic laser that’s 30 cm below to help with any kind of rotational and setup of the patients. Additionally, also keeping our therapists in mind, we have an Argo lift for patients not to grab onto therapists, but to grab onto that device. And we have two dual in-room monitors looking at ergonomics. Depending on which side of the couch you’re standing on, you can still see the monitors. That being said, there’s a lot going on in this room. We are not typically putting in any of those kind of skylights or those different items. When we first implemented the track, it was hitting the DoseRT camera, and then when we do treatment, part of our procedure is to keep that Argo lift in a different area.

I will also say that DoseRT is susceptible to red wavelength light, so we have green lasers, and I highly recommend that for anybody that is purchasing lasers. We’ve tried blue, but green is really great. Red would have an impact. Additionally, we had installed light filtering on each of these can lights, and it’s kind of hard to see, but there’s a little bit more purplish light from this light, whereas this is more white. This came from the same manufacturer, but different batches of the UV filtering. And so what we needed to do is really work with getting the filtering that worked the best for our implementation.

And like I said, we have the garden, so you can kind of see a little bit of snow. We have some flowers. When we originally put this vault in, we did purchase AlignRT, and the calibrations for that system worked really well. We were not thinking about DoseRT at the time, several years ago, and it was pre-DoseRT. And so we have now blinds that we bring in during treatment with DoseRT because you really do need a dark environment for that treatment, but it is a feature that we still have in the room and utilize. But I’ve had to modify our workflow a little bit just because of the ambient lighting and learning things from the installation from that perspective.

Some tips and best practices: you always want to turn off the ODI. That’s part of the procedure for us. It is not something that our therapists always do, but that would be key in terms of setup. I think looking at some of the site visits and what folks at Atrium and others are doing with setup of patients using postural video, I think really helps with that. Using AlignRT for virtual SSDs, the ODI is not necessarily needed. We do have some older vaults that have red lasers, but that would be problematic with the DoseRT.

The other element for us is there’s elements in the VisionRT system and the DoseRT system which are great. We have not only one set of in-room monitors, we have dual in-room monitors. So we had to install a switch to turn off the power to these devices to darken the room, really to get the best possible DoseRT images. Incidentally, as we look at adoption across many of our sites, some of our therapists who do rotate from a system perspective don’t like how necessarily dark the vault has to be for DoseRT, but there is a future solution for that that I will touch upon.

So probably the most important and best slide in my presentation in terms of DoseRT is with the ODI. If you don’t turn it off and you have the white light ODI on during treatment, this is what DoseRT looks like. It’s not great. What we have found, we have in-house clinical engineering support that also utilizes third-party equipment, is that there is a green light ODI which is now becoming more popular. And as you can see very faintly, that is how it appears on the DoseRT camera, and you can kind of see the line. So it doesn’t obscure the lighting as much as the white light ODI. So if I were going to get DoseRT, I would really try to have the therapist get in the practice of turning off the ODI. But I also think that this green light ODI is a great solution.

For us, early IT engagement is critical to export DoseRT reports to the network. We review it the first fraction and weekly, and a lot of times we’re using it for bolus placement and verification. Another thing that we have struggled with and look at and continue to think of is our workflow design. At one of our sites, we have all of the Varian control console equipment in one area with a therapist that we call the driver. The second therapist is really looking at the VisionRT system here, and we have the DoseRT monitor above. So they’re looking for monitoring the patient, collisions, and DoseRT information during treatment.

I think it’s very important to do some early site visits with folks that have DoseRT. I know Advent has some DoseRT systems here. And I think we had very good vendor support during our go-live. Staff feedback and buy-in is critical, and it’s a team project because, as I said earlier, some staff may not want to be in the room that’s completely dark, and you may need to make some modifications to your room and to the lighting of your room.

And presenting some clinical cases, and I think a number of folks have presented theirs as well. This one was particularly interesting and important to us with a breast treatment, a four-field breast treatment with bilateral breast being treated. So we really couldn’t have the patient turn their chin to one side or the other. We did see that during our treatment, and our therapist stopped the treatment, that the chin went into the field. So this is really the power of the system that I think you’ve seen from other institutions as well as something that we can share that we really think that it does show you quite powerfully and simply when you’re having an issue.

It does also change the way in which we evaluate plans. We have had some plans where there’s some chin dose, and I think some of the other speakers have spoken to it before, where we were looking at the 50% isodose line, but really not looking at the 20 or the 10, and can really be better planners in terms of our radiation delivery based on the input that we receive from DoseRT.

Some of our clinical cases and some of the strengths of the system: verifying the appropriate flash and setup. We see a great benefit in bolus placement as well. And we do see some limitations, particularly with skin pigmentation, camera FOV for certain sites, as well as modulated fields. So looking at the example here, we have a patient positioning error with the patient’s chin turned into the field with the correction that we could see, which quite simply, it just makes our treatments all the much better. You can see that here you have some dose, but the bolus is misplaced and then placed in the correct position.

Modulated fields can sometimes be a little bit noisy. Right prone breasts also have limitations due to the camera. And then also as we look at tattoos, that can be an issue in terms of really mapping the dose accurately. It is very highly sensitive, but requires a very good room setup. But oftentimes, the physicists, we look at the purple and we’re wondering, “Well, how much dose is it?” It’s really a relative, but not an absolute dose yet. And so it’s just a really good indication for something for us to focus on.

We see a lot of value in evaluating our SRS and SBRT plans in a side-by-side plan review versus treatment review, and looking at intrafraction motion and position variability. Ambient light and reflection is a problem, sometimes from Vaclock bags, bolus, the LINAC housing. I often joke I wish I could buy my LINAC in matte black because that would cause less issues. There is, and I’ll talk a little bit about it, an upgrade pending to improve light filtering. We are actually going to be receiving that in the next 45 days, so maybe at the next year’s talk, I could talk about that implementation as well. Because sometimes you get false positives from reflections that do not represent true dose. Some of the items in terms of the review really just display the date and time. I know that there is a software coming and an upgrade that’s going to display the patient name, which makes it a lot easier for review. So by the time you do get DoseRT and with the upgraded software, it will be better than it is today.

Talking about future optimizations, Cherenkov imaging and patient comfort: it really does require a very strong focus to lighting. I know when we’ve talked to our construction folks about it, lighting and where those air deflectors go is not always mapped out well, and we’ve stressed that it needs to be for the devices, really managing room lighting conditions to get the optimal environment, and to be able to clearly visualize and see dose. I am thinking that the BeamLight solution is going to be a vast improvement for not necessarily needing to darken the room and having the lighting all tied in to the VisionRT system. So I’m very grateful that that issue that we have seen is being resolved.

I think there are five things in radiation therapy that excite me. DoseRT is one. I think Dr. Zimmer was speaking about grid therapy is another, and spatially fractionated radiation therapy. I know we’ve had a couple cases in the past month. And just seeing all of the technology coming together, and really for the benefit of patient treatment is something that we continue to support, continue to be very enthusiastic about. And with that, I’ll take any questions. Thank you.