How soon will we have robot surgeons? Health care has gone remote lately, but in reality, most of it is fairly simple: video conferencing during Covid. Autonomous robotic surgery is more science fiction than science fact right now.
But just recently for the first time ever a robot surgeon at Johns Hopkins University performed abdominal surgery on soft tissue. Granted … it was on a pig, not a human … but STAR, or Smart Tissue Autonomous Robot, was a success. And that means there’s significant hope that robot surgeons are not only possible but a reality.
Support TechFirst: Become a $SMRT stakeholder
Unfortunately, it might be a decade or two before this is normal. But technology does advance quickly … and we need it to. Healthcare is unevenly and inequitably available both globally and nationally, and cheap, fast, effective robotic surgery would be a huge boost to health care outcomes.
Check out the Forbes story here, or keep scrolling for full video, the podcast, and a transcript.
TechFirst podcast: robotic surgery
Transcript: How soon will we have robot surgeons?
(This transcript has been lightly edited for length and clarity.)
John Koetsier: Okay, how soon will we have robot surgeons? Healthcare has gone fairly remote lately. In reality, most of it is very simple, we’re talking video conferencing during COVID. But just a week ago, for the very first time ever, a robot surgeon at John[s] Hopkins University performed abdominal surgery on soft tissue.
Now granted, it was on a pig … not a human, but STAR, or smart tissue autonomous robot was a pretty significant success.
Question is, how soon will this be available for humans? To chat about it, we have Dr. Tamir Wolf who’s the CEO and co-founder of the surgical intelligence platform Theator. Welcome, Tamir!
Dr. Tamir Wolf: Hi John, thanks for having me.
John Koetsier: Hey, it’s a real pleasure to have you. Thank you for taking some time here. Give us a scoop … what happened at Johns Hopkins?
Dr. Tamir Wolf: Well, it’s the first time that minimally invasive surgery was actually performed on live tissue.
It was a pig … and it was very interesting, because they were able to suture the intestines of the pig by leveraging the capabilities of autonomous robots. So this was the first time that this has been done. It is predicated on work that was done in 2016, or published in 2016 if I believe correctly, and it takes us like one step further in the evolution of autonomous robotic surgery. So, quite exciting.
John Koetsier: Now, so this is minimally invasive — so there’s a small incision, instruments are inserted, the operation happens without a large incision happening — and if I’m not mistaken, they picked this because doing soft tissue, the intestine, that’s one of the hardest surgeries to do, right, to stitch that together. Is that correct?
Dr. Tamir Wolf: Yeah, so there are aspects of dexterity and just the ability to take soft tissue and suture it … it is difficult. I’d say it’s not the most difficult part of a procedure because the most difficult parts of the procedure are actually the cognitive aspects, and I guess we’ll talk a bit about that, but I think like from a functional standpoint of things that surgeons do repetitively during surgery, this is definitely something that takes us a bit forward.
John Koetsier: Let’s talk a little bit about what it takes, what you need to make this happen.
Clearly there’s a hardware component, there’s a software component, there’s some intelligence required. There’s a lot more than that as well, there’s regulation, there’s ethics … what do you all need to make something like this happen?
Dr. Tamir Wolf: So that’s a loaded and it’s a huge question. Obviously there’s so many different aspects to this.
I think, if you take a look at the evolution of robotic surgery and compare it to what’s happening like in the automotive industry, for example, you have various steps from like no automation at all, to robotic assistance, to partial automations like ADAS, you know, with cars. And then after that, there are like, conditional automation where the vehicle starts understanding its environment. And then there’s the high automation aspects where that’s really the decision-making process, and ultimately, full automation.
I think where we are today, or where this work at Hopkins takes us forward, is around the partial automation aspect. We have the capability, or like this is the initial proof of concept of the capability to automate tasks.
And so I think that’s where we are in the evolution of robotic surgery, which, let’s face it, today is kind of a misnomer because it is actually a human surgeon that’s leveraging a huge device, moving a joystick and actually doing the procedure. So it’s not really robotic and, you know, robots today, they’re not that intelligent.
And so it’s the human behind it that is intelligent, but as more and more functionality transitions to the robot from that human know-how, then we start tackling a lot of these issues, like the ethical issues, like the regulatory issues, like other issues all around it. So, I think it’s very gradual, like, I think it’ll take quite a bit of time. We’ve been talking about autonomous vehicles for… I dunno, it seems like ages and that’s taking time.
And so I think we’re moving in the right direction and this is a huge leap, but it’s still like a small step when we’re looking at the broad scheme of things.
John Koetsier: One thing we chatted about just before we started recording here was who’s responsible for a screw up.
I mean, I remember, I think it was a few months ago, maybe half a year ago, somebody was using the self-driving functionality on their Tesla and they requested the car to come to them — I think it’s called Smart Summon, you summon your car from the parking lot, it’s supposed to show up, you know, and you’re kind of like you step in, the chauffeur brought your car, the valet brought your car, you step in and go — and unfortunately the car crashed on its way from the parking lot to the pickup spot, and they were going like, ‘Hey, who’s responsible? Is Tesla going to cover that?
Does my insurance have to cover that? And who’s responsible for a screw up here if the robot surgeon messes up?
Dr. Tamir Wolf: It’s a great question and I think, you know, we have at least 10 or 15 years until we have autonomous surgery. And I think, leading up until then, and even then, I think there’s going to be like a significant component where everything is really supervised by a surgeon.
And what the robot does is actually supports the surgeon in a variety of tasks, and hopefully decisions down the road that help minimize variability in the way that surgery is performed, because there is tremendous amount of variability today. And using devices and software-like robotic surgical equipment, we can actually minimize that type of variability.
So I think we’re still a long way from robots doing everything on their own, and so accountability will likely still be with the surgeon who is responsible for performing the operation.
John Koetsier: I think that makes sense. Let’s talk about what that timeframe looks like. We’ve seen autonomous surgeons do stitching, for instance, and theoretically the report is they’ve outperformed humans at doing that.
When do you think we’ll start seeing robotic surgery — and we won’t call it autonomous, let’s say maybe let’s make that two levels: robotic surgery on humans and then autonomous robotic surgery on humans — what do you think is the timeframe here?
Dr. Tamir Wolf: I think it’s going to be evolution of like the robotic platforms where initially they assist in very discreet, specific parts of the procedure where we need, like the surgeon needs to do something that requires a lot of repetitive motion, and that for them might take a long period of time and not be as precise as what can be done with a robotic platform.
But it’ll only be for like specif— in my mind, it’ll only be for specific portions of the procedure. A huge part of what makes or breaks a procedure in the operating room is actually the decision-making process. It’s how you interpret things and what you do when you encounter something that just happens.
And I think we’re a bit far from that — it’s actually, at Theator, we’re trying to start and gather all that, like we call it the ‘surgical intelligence’ because it really is the brains and the know-how that surgeons have today that is scattered all over, and we’re trying to start understanding what that is and codify it so that ultimately one day we’ll be able to just fuel it into the robotic platforms so that there is this cognitive aspect, but it’s going to take a bit of time. Yeah, go ahead.
John Koetsier: It’s almost like you’re saying that every human being is just slightly different and every surgery is just slightly different. I mean, if you talk about robots that are assembling products, each product is exactly the same size, the components are exactly the same dimensions every single time — or else there’s been a screw-up, and if so, then the entire production line is probably shut down — but humans come in different sizes [laughing] … and I suppose that happens inside as well as outside.
Dr. Tamir Wolf: Definitely. So humans come in different shapes and sizes, and surgeons do too. Surgeons have different capabilities.
Surgeons, like the way surgery is taught today is an apprenticeship model. Every surgeon, even within a specific department today, has different background, has different exposure, has different experiences. And so there is this aspect of variability, and I think this is where, you know, this is the real promise of robotic surgery down the road: you minimize that variability and you’re able to actually tackle disparity by understanding best practices and then leveraging the robotic platform to actually perform in the same way … but you need to know what best practices look like, and again, that’s different between patients, just like you mentioned.
John Koetsier: How can surgical intelligence, like what your platform provides, feed into the AI and the machine learning that the robots will need?
Dr. Tamir Wolf: So, it’s all about the data, right? You need to gather a ton of experiences from around the world in order to be able to provide best practices and like the optimal outcomes, ideally, for a wide and very different patient demographic.
And so what we’re doing today is actually starting to aggregate that information, and we’re gathering experiences of thousands of surgeons, tens of thousands of procedures, and the idea is to have a wide variation there. And by doing so, the idea is to leverage the power of computer vision and artificial intelligence to not only start analyzing what is going on in the procedure, but derive insights from that.
So we basically leverage AI to put structure around surgical videos to understand what is going on there, and then we link it to the patient that goes into surgery, to the outcomes post-operatively, and to the surgeon that is performing the procedure. And by doing that, we’re, like the objective is to really understand what best practices in a specific scenario look like.
And then ultimately the idea is to be able to feed that type of logic into the robotic platforms so that when something happens during surgery they can actually tackle it.
Because when you’re doing surgery, even on a pig, you know, it’s one thing, but when you’re starting to operate in humans, it’s not only about passing a suture through the gut, it’s what happens if/when that suture nicks a blood vessel? What does the robot do then? How does it tackle that?
John Koetsier: Mm-hmm.
Dr. Tamir Wolf: There’s a lot of decisions and in surgery, like one bad decision, it’s like a snowball effect … it leads to another, another, and another. And so the ability to not only identify structures or be able to pass a suture, but actually to think the way that humans think, I think is paramount here. And with human surgeons — think about it — it takes decades for them to perfect their craft. It’s all about honing in these skills of decision-making, and this is what differentiates a good surgeon from someone who’s not as good.
John Koetsier: Right, right. It makes me think, what are some of the worst words you can hear from a doctor or a surgeon, right, and ‘That’s odd’ [laughter and crosstalk]…
Dr. Tamir Wolf: Hopefully you’re under anesthesia when that’s it, but yeah.
John Koetsier: [Laughing] Exactly. So peer into your crystal ball for us here. When do you think we will see autonomous surgeons operating on humans, perhaps under supervision, but perhaps not under direct, immediate, one-surgeon-to-one-robot supervision. How long do you think that will take?
Dr. Tamir Wolf: I think it’ll take a couple of decades, at least.
I think it also depends on the Space Race, because where a lot of this might be valuable are like trips to Mars, for example, where you can’t bring a specialist for each and every procedure, and you want someone or, in this case, something that has tremendous data in the back of their proverbial mind to be able to do things.
But I think we still have a ways to go.
John Koetsier: What will that do — assuming we get there — to the availability and cost of healthcare?
Dr. Tamir Wolf: One of the reasons I founded Theator is to tackle variability in this space.
Like I have a personal story where I diagnosed my wife and my previous boss with appendicitis while living in New York, within a span of several months, and I took them to the hospitals seven miles apart and the approach to treatment and treatment were very, very different.
With my boss, it was like a near-death experience, even though I brought him already with a clinical diagnosis … and again, one error and mistake led to another and complications and protracted stay. And on the other hand with my wife, luckily for me, 12 hours in and out. And so, this is like New York City, it was 2015 at the time, you know, two hospitals that have amazing surgeons. So why is that?
And when you think about it, there is tremendous variability because surgery is an apprenticeship. And I think being able to gather all this data, identify best practices, and drive them into a platform like this would go a tremendous way towards alleviating variability in the way that surgery is performed, and also allow better access to optimal care to huge populations that today simply do not.
Because today, you know, whether we like it or not, there’s this notion of like where you live determines if you live, just because surgeons are different. And so this is something that—
John Koetsier: And sometimes there aren’t surgeons where you need them to be. There are entire countries — whether that’s India, whether that’s other countries in Asia or Africa, even China, even, frankly, in the United States, perhaps the richest country on the planet — there are many locations, rural locations where you simply don’t have access to the doctor, any doctor perhaps, but certainly in many cases the specialists you need and the surgeons you need.
Dr. Tamir Wolf: Exactly exactly. There was a recent article that also came out in the New York area that talked about hysterectomy or resection of the uterus and showed that if you’re at a higher socioeconomic status you go to the specialist, and if you’re lower socioeconomic status then you go to the doctors that have less experience, in the same healthcare system.
And so that is the type of thing that even in the United States, we, I think we’re tackling, and I think there’s tremendous promise in a platform like these surgical robot platforms that we’re working towards.
John Koetsier: Wonderful.
I think it is a fascinating future that we’re looking towards in healthcare, because we have the convergence of so many different things. We’ve got so much data. We’ve got intelligence in the form of AI. We’ve got robotics. We’ve also got so much data that we’re collecting every day from wearables that we’re wearing, data that we’re collecting from apps that we’re tracking what we’re eating, apps that are tracking our exercise … and that’ll get more and more sophisticated as we go as well, looking at our heart and other vital functions, and certainly this one right now [pointing to Apple Watch] can tell the oxygen level in my blood, and what are we going to have in five years, right? Is it going to look at blood chemistry on a regular basis? Am I going to get something inserted? Who knows, right?
But the AI platforms that are going to be required to synthesize all this data — because frankly, your general practitioner doesn’t want to hear it [laughing] he or she has way too much going on to look at your Apple Fitness data or something like that — but synthesizing all that, bringing that to bear on an ongoing basis but also in the moment when you need specific care, invasive care, is — and also the recuperation process after that — it is a fabulous, amazing, wonderful future that we’re working towards, and I want to thank you for this time.
Dr. Tamir Wolf: Thank you. Thank you, John. Happy to be here.
TechFirst is about smart matter … drones, AI, robots, and other cutting-edge tech
Made it all the way down here? Wow!
The TechFirst with John Koetsier podcast is about tech that is changing the world, including wearable tech, and innovators who are shaping the future. Guests include former Apple CEO John Scully. The head of Facebook gaming. Amazon’s head of robotics. GitHub’s CTO. Twitter’s chief information security officer, and much more. Scientists inventing smart contact lenses. Startup entrepreneurs. Google executives. Former Microsoft CTO Nathan Myhrvold. And much, much more.
Consider supporting TechFirst by becoming a $SMRT stakeholder, and subscribe on your podcast platform of choice: