Augmented Reality for Multi-Needle Composite Liver Tumor Ablation

In this episode of the SIO Corner podcast, Andrew Kolarich, MD, Johns Hopkins Hospital, Baltimore, Maryland, interviews Katerina Lee, MD, University of Pennsylvania, Philadelphia, Pennsylvania, about her study regarding the use of a smartphone augmented reality guidance system for composite ablation margins in phantom models.

In this study, the smartphone augmented reality guidance system performed better than conventional CT guidance, particularly with less experienced radiologists.

Transcript:

Interventional Oncology Learning: Welcome to IOL Radio, the podcast for I /O Learning, a digital publication that covers the latest advancements in interventional oncology. This podcast episode is part of the SIO Corner, a collaboration between IO Learning and the Society of Interventional Oncology. Our guest host today is Dr Andrew Kolarich, an interventional radiologist and member of the SIO's Publications Committee. Dr. Kolarich's guest is Katerina Lee, a fourth-year medical student at McGovern Medical School at UT Health in Houston, Texas. [Editor’s Note: Katerina Lee has since graduated from medical school and is currently a Research resident with University of Pennsylvania Diagnostic Radiology.]

Katerina is a former research fellow of the Medical Research Scholars Program at the [National Institute of Health] NIH, where she participated in numerous research projects. She is here to discuss her recent Society of Interventional Oncology abstract, Augmented Reality (AR) for Multimodal Composite Liver Tumor Ablation, identifies tumor at risk for under-treatment.

Andrew Kolarich, MD: Hello, this is Andrew Kolarich, one of the interventional radiology residents at Johns Hopkins Hospital in Baltimore, Maryland. I am here with Katerina Lee. She's a fourth-year medical student at McGovern Medical School in Houston. I'm talking to her today about a really interesting abstract that she presented at the Society of Interventional Oncology, titled Augmented Reality for Multimodal Composite Liver Tumor Ablation Identified Tumor at Risk for Undertreatment. Welcome, Katerina.

Katerina Lee, MD: Hi, thank you for having me on this podcast. I'm super happy to talk about my AR project here.

Dr Kolarich: Starting off, what was the purpose of your study? Just give us an outline of how you designed it and what endpoints you were measuring.

Dr Lee: The purpose of our study was to evaluate a smartphone augmented reality guidance system for thermal ablation of large tumors at risk for undertreatment. To test this, we developed gel phantoms with large CT-visible tumors inside. And we had 4 operators of various years of experience who performed iterative ablations using both the AR guidance and CT guidance. And then we compared the composite volume of ablation for each of the guidance methods.

Dr Kolarich: It sounds like you had 2 separate types of targeting methods that you were using, one using augmented reality and the other CT, which I think clinically CT is certainly more common now, but that may not be the case. What was your patient population or sort of tumor type or size range that you were working on?

Dr Lee: All of our experiments were performed on gel phantoms made by one of our team members. These gel phantoms had a tumor inside, which had a different house fuel value than the surrounding tissue, allowing us to visualize the tumor on CT. We specifically designed large tumors because tumor size is known to be associated with disease progression or recurrence.

With large tumors, obviously, it's more difficult to achieve adequate ablative safety margin, which has been proposed to be from 3 to 10 mm. And the safety margin is also associated with better patient outcomes, so navigation systems that allow the operator to better visualize the tumor in real-time during a procedure may enable a more complete ablation of large tumors and subsequently result in better patient outcome.

Dr Kolarich: Absolutely. Clean tumor margins are critical, in particular, in folks that can't get surgery. Interventional radiology and ablation; that's one of the curative options. It's a big, big topic. And trying to get these folks extended survival time, get them potentially to transplant, all of these things hinge on being able to treat tumors. And sometimes those big ones we know are challenging. So that definitely makes sense.

If you're able to outline your results, are there any sort of adverse events or difficulty with treatments during the process?

Dr Lee: With our AR guidance system, the mean percent coverages of large tumors are greater than that of the CT-guided freehand ablations. It was, in particular, interesting that for the AR guidance method, there was no difference in mean percent coverage of tumors among operators, despite their various training experience, whereas for the CT-guided freehand ablations, there was a significant correlation between years of experience and mean percent coverage.

What we notice is that when using AR guidance, there was less overlap of these ablation zones, compared to those from the standard freehand CT guidance, leading to more coverage of the large tumor with the AR guidance system. We think that this really speaks to the potential for navigational technologies in general to support standardization of procedures. This potentially could make it more likely to receive the same procedural accuracy or care, no matter where it's done or what level of experienced operator is available.

Dr Kolarich: That is really, really interesting. And I think the perspective here that's useful for me, as someone who's now in their more senior level of training and doing more of these ablations, is getting good coverage can be challenging in large tumors. You often have multiple probes, depending on if you're doing microwave or cryoablation, and kind of choosing that.

We do a pretty high volume of ablation that has really picked up over the last 2 years at Johns Hopkins, and I've heard that across the country. But that ability to understand your coverage under CT and maybe having better coverage standardization would be huge, because a lot more interventional radiologists are going to be doing this, not just at large academic centers, but in community practices.

Did anything surprise you about your study results?

Dr Lee: It was a bit surprising, although explainable, to see the correlation between percentage of ablation and years of experience with CT -guided freehand ablation, versus the lack of this correlation for AR -guided ablations. This lack of correlation between ablation outcomes and years of experience has been previously reported in other studies as well. While using other navigational technologies.

And additionally similar, more standardized, normalized, and reproducible procedures with shortened learning curves using other navigational technologies for ablation, such as fusion and stereotaxi have also been reported. It was nice to confirm similar findings in our study as well.

Dr Kolarich: Interesting, so it sounds like if a person can understand how to use the augmented reality, the learning curve might be lower than our traditional sort of CT guidance that we've been using currently. Very cool.

Dr Lee: Yeah, potentially, yes.

Dr Kolarich: We've been talking about the augmented reality. That is not something I'm super familiar with, on a clinical level. Can you describe that novel technology and give a brief overview of how it works?

Dr Lee: The novel technology involved is the augmented reality guidance that we implemented in our project. As you probably already know, augmented reality has recently been introduced into minimally invasive surgical fields for use in image-guided needle procedures on a variety of platforms, including smartphones and goggles. Essentially, AR enables superimposition of pre-procedural images onto real-world patients, allowing the operator to use digital information and reality perspective simultaneously in real time.

And other benefits of the system include reduced learning curves and less radiation exposure. But using AR procedures requires accurate 3D-modeling of patient anatomy prior to the procedure and precise registration, which is basically mapping and tracking during the procedure. With accurate registration and tracking, AR basically has the potential to overcome some of the limitations of standard 2D image-guided procedures.

And what's more unique about our approach is that we use smartphones and smart pads instead of goggles, which avoids the ergonomic difficulties of wearing goggles for an entire procedure. Most other AR systems rely on goggles alone. For us, we take over the gyroscope, the accelerometer, and the camera of the smartphone for smarter procedures.

Dr Kolarich: That is really, really cool. That component of mapping and pre-registering, it's something that radiologists in general are going to have an upper hand doing because part of our training is so heavy on MRI CT, understanding 3D relationships of things, even ultrasound. That's really really cool. And I'm looking forward to maybe in a couple years using my smartphone to guide myself instead of just my fluoro CT.

Are there any future studies or potential future studies?

Dr Lee: Future studies may focus on minimizing some of the limitations that we notice during our trials. This includes the effect of the lighting of the room on accuracy of tracking. We noted that angles of brightness and shadows can affect how the smartphone captures and tracks the phantom and the ablation needle. Inaccurate tracking obviously can lead to air latency, jittery imaging, and delays in procedure. This would also be critical when evaluating the translational value of this research. Real patients are going to move and there will be internal dynamics such as breathing and the movement of the digestive tract. That will make accurate real-time tracking even more critical and difficult. Focusing on enhancing its accuracy will be very important in moving forward.

Ultimately cool technologies like this one are helpful to augment but not to replace the physician.

Dr Kostarich: That's a huge thing and something I learned from more senior interventional oncologists that I worked with, particularly taking those individual patient factors and doing things in real-time. I have a very wise older interventional oncologist I work with and every time I complain about breathing motion, he reminds me breathing motion is critical because that means your patient's still breathing. And that’s just part of it, right? So that's interesting.

Any final parting thoughts about your project or anything that you're looking forward to in the future in this space?

Dr Lee: Right now is a really exciting time for research and radiology, especially with all the virtual reality, augmented reality, and AI projects going on. I hope this type of research will continue to develop and I'm super excited to see how state-of-the-art technologies like our project can be eventually implemented clinically, as well as how they can be further developed to meet the needs of both physicians and patients.

Dr Kolarich: Absolutely. Awesome. Well, I appreciate your time, Katerina. Again, I'm Dr. Andrew Kolarich. I'm one of the interventional radiology residents at Johns Hopkins. And this has been IO Learning Podcast, looking at augmented reality for multi-needle composite liver tumor ablation at tumor risk for undertreatment. Thanks again, Katerina.

Dr Lee: Thank you.

IO Learning: That wraps up this episode of IOL Radio. To hear more conversations about the latest advancements in interventional oncology, be sure to visit the podcast page at IOLearning.com.

Source:

Lee KH, Li M, Varble N, et al. Smartphone Augmented Reality Outperforms Conventional CT Guidance for Composite Ablation Margins in Phantom Models. J Vasc Intervent Rad. 2024; 35(3):452-461.E3. doi:10.1016/j.jvir.2023.10.005