Reimagining Surgery Through AI and Computer Vision Technology With Tamir Wolf, MD, PhD, and Waqas Haque, MD, MPH

In this episode of Exploring Artificial Intelligence in Oncology, Dr. Waqas Haque sits down with Dr. Tamir Wolf, Co-Founder and CEO of Theator, a company pioneering next-generation AI and computer vision technology to automatically capture and analyze surgical data in real time to generate actionable insights and improve surgical outcomes.

This episode is also available as a podcast. Click here to listen! 

Waqas Haque, MD, MPH: Good morning, everybody. Welcome to Oncology Data Advisor’s monthly podcast on our AI and technology theme. We’re joined by our guest today—Dr. Tamir Wolf, who’s the Head of Theator, which we’ll be talking about today. Thanks so much for joining today, Dr. Wolf.

Tamir Wolf, MD, PhD: My pleasure.

Dr. Haque: I know you’ve had an interesting background before being at Theator. You’ve done a lot of work in health entrepreneurship, you’re a Navy SEAL, you did an MD/PhD, a lot of cool things. Tell me a little bit about your pathway before coming to the company, and then how it was founded.

Dr. Wolf: Again, thanks for having me. I’m always delighted to talk about things that are interesting to the clinicians out there, especially those that are also interested in embedding themselves in technology. To talk a little about my path, I’ve always been an outlier. I did an MD/PhD and was one of the first in my class to do that. I went to med school as an ROTC and ultimately found myself in the Navy, where I spent quite a bit of time in Israel. I needed a break afterwards and found myself in tech. I left clinical practice and got my MBA in Boston.

I then decided that I wanted to make an impact, not on an individual level, but I wanted to have impact on a broader scale. That’s what led me to think about, “All right, where’s the unmet need and what can we do to tackle it? I ultimately found myself at Theator after realizing that there’s significant variability in surgical care specifically—in healthcare in general, but in surgical care in particular, just because of the siloed nature of residency, fellowship, and practice. So, I dove all in.

Dr. Haque: I was reading online that you had some people close to you who had been recommended to different hospitals for a certain procedure, but then they had completely different outcomes.

Dr. Wolf: Right, and even when you take a look within departments, you see tremendous variability in how surgery is performed. For hundreds of years, folks have been talking about surgery as an art, but I think there’s a lot that we need to do. I think it’s science, honestly. There’s an art form to it, but I think that we really need to do a much better job of standardizing the key elements. At Theator, we’re trying to unearth and surface what is actually going on in the operating room and then ultimately codify surgery and skill access to high-quality, low-cost care.

Dr. Haque: So, you’ve had this really cool background in healthcare entrepreneurship and different kinds of settings. You got involved in Theator. How do you try to integrate AI to surgical outcomes? What kind of things are you looking at, and what’s an example you can talk about that shows the success so far?

Dr. Wolf: When we started out, my idea was that we need to help guide surgeons in real time. What I realized very quickly is that in surgery, which I think very different from other domains in healthcare, there is very little consensus on what best practices even look like. In order to guide folks on what to do, we first need to understand what good looks like. Then it became apparent that the operating room (OR) is pretty much the last silo of lack of transparency. You’ve got cameras everywhere, all over the hospital. You’ve got data coming from everywhere. But the OR is the silo where, for a very long time, what happens there stays there.

The foundational layer for everything that we’re doing is, number one, capturing information and data. When you think about data from the operating room, you can break it down into two big buckets.

One bucket is slapping on a couple of cameras within the room and taking a look at team dynamics and turnover, etc. The other bucket is analyzing operative performance. We went down the second route. It’s very easy to put a couple of cameras in the room, technologically speaking. It’s very difficult culturally to ensure that the folks don’t disconnect them from the wall. When we are talking about intracorporeal surgical footage, a lot of surgical procedures are done minimally invasively, whether laparoscopic, for example, or robotic, so the visual guidance is already there.

What we do is we routinely capture intracorporeal performance in order to start building an entire universe around that. Even this is a heavy lift within the world of surgery because up until now, no one’s ever done it. But we believe that what happens in the OR matters, so we enable routine capture of intracorporeal surgical footage. We then connect it to the patient and to outcomes in order to understand what happens in the OR and how it impacts outcomes, which I think is important, and we’re doing for the first time.

I’ll give you just one example. Critical view of safety is something that general surgeons have been talking about for 20 years in laparoscopic cystectomy in gallbladder surgery ever since they invented the approach. Yet no one’s really been able to identify whether or not this is a best practice that’s actually adopted in routine real-world care. We have the capabilities to leverage algorithms in a very scalable way to identify in each and every single procedure, thousands on a weekly basis, whether or not this specific element which has been shown to lead to better outcomes is being incorporated into the procedure. This is an example of, for the first time, incorporating into real-world workflow the ability to identify whether best practices are being adopted on our way towards real-time decision support.

It’s just like autonomous vehicles, still a bit away, but this is how we start understanding whether best practices are adopted or not. We’re actually able to demonstrate in departments that there’s a cultural buy-in. We’re able to demonstrate an improvement in adoption of this and other best practices over time and reduce operative time, which is extremely important, especially in today’s environment. There’s a lot of value from an operational perspective, obviously from a clinical perspective, and it’s also valuable for the patient by impacting the finances, which is extremely important to talk about as well.

The other thing I would mention is that the video is really the ground truth of what happens in the operating room. You can leverage that to do a variety of things, including everything that I talked about from quality improvement, patient safety, avoiding never-events, the operational aspects, but you can also leverage video in order to improve the way things are being done. For example, automation of the operative report is something that we’re right now working on, and ensuring billing integrity is a proof of concept that we’re just now working on. There are a variety of elements both surgeon-facing, department-facing, and also healthcare institution–facing that are all based on this notion of routinely capturing procedures that are being done and leveraging the information for a variety of unmet needs.

Dr. Haque: Wow, that’s so amazing. Talking on about the art versus science of the surgical OR is really, really interesting. For medical oncology, you have this option to just randomize patients, do different arms, look at overall survival, and be able to get a data comparison based on characteristics, but in the OR, there are hundreds and thousands of variables that you’re sometimes looking at. I got a first-hand look at that in medical school, and I’d be interested to see if different surgical attendings have different outcomes for patients. That’s really cool. I’m also curious to hear how the Theator’s platform could be used to improve graduate medical education for residents or surgical trainees.

Dr. Wolf: We’re working very closely with a few departments at Mayo Clinic. We have residents and fellows that spend a lot of time on the platform. They have the ability to take a look at a procedure and then asynchronously interact with one another, provide feedback, and receive feedback. There are numerous ways from that perspective. I think the Gen Z and future generations of surgeons are all going to be extremely tech-savvy. Just like everything in life, we’re consuming video as a key form of information. It’s very similar with being able to view these procedures and interact with one another. You don’t need to be in the same room; you could just do it asynchronously. That’s something that we’ve seen is extremely powerful.

Theator is being incorporated into the weekly journal clubs that residents and fellows have and a lot of other things. They have the ability to, within a second, get a specific moment in time in a video that is automatically identified for you. You don’t need to run through hours and hours of footage; all the different steps and advanced key milestones in a procedure are automatically identified. The ability to jump to that and interact on a specific moment in time facilitates something that’s a lot more user-friendly and easier to do. Residents and fellows love the platform.

Dr. Haque: That’s awesome to hear. Now, just shifting gears from Theator’s platform, tell us a little bit about yourself and your entrepreneurial background. You’ve spoken before about the importance of company culture driving success. What are some of the key ingredients that are needed to be successful in the AI healthcare startup space?

Dr. Wolf: I think timing is important. That’s something that you can’t really control, but today there’s a lot of willingness to accept these types of technology from healthcare systems. I think that’s extremely important. Selling to healthcare systems is not easy, so it’s something that can be extremely helpful. As for the ingredients from a company perspective, I think it’s a good combination of clinical domain expertise and technical know-how. There are a lot of interesting technological solutions, but how do they really answer an unmet need that is true and real? That’s only something that you can figure out if you have the right individuals who can ask the questions. I think that combination has been something that’s really powerful for us, because the way we’re looking at it, we’re building a variety of solutions that can incorporate within existing workflows. It’s not like a huge cultural paradigm shift, even though it’s cutting-edge. The ability to do that relies on having those two components.

Dr. Haque: Just to build upon that, you mentioned domain expertise. With all the change going on in the world of AI in medicine, with all the daily updates, all the new trials, all the new studies, what’s your information diet like? How do you keep up with new stuff and new findings in the literature and press?

Dr. Wolf: From a technical perspective, there are specific areas and resources that you go to, but honestly, we’re the ones that are creating a huge delta in our domain. I think more important than this algorithm or that algorithm is really understanding the basic issues and problems that you’re trying to solve. Again, technology is cool, but if it’s not providing value, it’s meaningless. I think being embedded within the healthcare system and understanding the exact pain points of a surgeon or clinical leadership or administration as it comes to surgery, ultimately serving patients, that is the most important element. Then everything else is just on top of that.

Dr. Haque: Then just to add onto that, how would you recommend that oncologists or clinicians who are in practice should they learn about AI? How can they incorporate it into their practice? What are some tips you would have for those people?

Dr. Wolf: AI is a buzzword right now. The way I would look at it is to really think about the issues that we as physicians or oncologists encounter on daily basis. What are the problems? What’s an unmet need that we identify? Then from there, try to look for solutions that can actually help us from that perspective. Whether it’s in the diagnosis of an illness or whether it’s in a workflow that we want to enhance and improve, AI can do a lot for us. For example, it really is an enabler of automation and doing things at scale. But I think it’s important to come from the angle of “What is the problem that I’m trying to solve? Are there solutions for it?” If not, let’s start diving into that rabbit hole and figure it out. If we can come up with a solution, I think that’s the best way to approach something like this. It’s not so much a hammer looking for a nail, but it’s the other way around.

Dr. Haque: Awesome, thanks for sharing that. Final question, do you have any recent partnerships or working developments at Theator?

Dr. Wolf: Like I mentioned earlier, we’re working very closely with Mayo Clinic, University of Miami, and other institutions. A lot of the things that we’re working on right now are continuously expanding on the offering of analysis of routine video capture. We not only capture the intraoperative video, but we also connect the dots with the medical record, with the idea that the magic that happens is really when you marry intraoperative performance and postoperative outcomes. But like I mentioned before, there are multiple areas where we’re now diving a lot deeper.

For example, imagine a world where you don’t need to write an operative report, but it’s automatically generated for you. It’s going to be a lot more accurate and introduce a lot less bias than if it’s a cookie cutter template, which is most often the way things are being done today. It completely revamps and changes the way things are done. Right now, we’re working with a select group of partners to actually push this paradigm shift in the world of surgery. I think video is the ground truth, as I mentioned several times on this call, that allows us to really generate things in a very, very different way.

Dr. Haque: Thanks for sharing, Dr. Wolf. Thanks to everyone for joining our podcast today. I think this is a really cool example of someone using their medical expertise, noticing things in the real world, and trying to drive structural change in healthcare. All the best with the company, and thanks so much for

Dr. Wolf: Thank you.

About Dr. Wolf and Dr. Haque 

Tamir Wolf, MD, PhD, is the Co-Founder and CEO of Theator, a company pioneering next-generation AI and computer vision technology to automatically capture and analyze surgical data in real time to generate actionable insights. With a rich background in health entrepreneurship, Dr. Wolf leads a team of over 50 people in Palo Alto and Tel Aviv in building a surgical intelligence platform that aims to improve the health of patients, the performance of doctors, and the overall level of hospital care.

Waqas Haque, MD, MPH, is a Hematology/Oncology Fellow at the University of Chicago. He recently completed his Internal Medicine Residency at New York University (NYU) in a Clinical Investigator Track. Dr. Haque has balanced his patient care work with a variety of research projects. During his fellowship training at the University of Chicago, he plans to further his work in innovative clinical trial design, value-based care delivery to cancer patients, and clinical investigation.

This transcript is edited for clarity. Any views expressed above the speakers’ own and do not necessarily reflect those of Oncology Data Advisor.