Will AI Robots Perform Surgery Better Than Humans?

32:22And this surgeon will have shown you because they perform surgeries and they deliver the best possible care day in, day out. And so that you can learn. But autonomous systems and robots are not people. And the algorithms that drive them, even though we call it AI and that gives you this idea that maybe they're similar to people because I try to train them on demonstrations of real people. they just fail in very different ways right like they fail in ways that people would probably never fail they go in a little bit in the wrong direction and afterwards like if you don't develop them correctly they will get stuck because they now see a situation that they have never seen during training because the surgeon just never moved there right like the surgeon knows this is not where you go you go elsewhere but if if the system makes a mistake for various reasons because somebody bumped into it because the sun didn't play ball, because the algorithm just didn't get it right and it moved to a slightly wrong location.

33:27How do you allow the system to recover from situations that in the real world you just don't necessarily observe? And so this, I think, is part of what we're building at EnerLogic also is the ability to synthesize these types of scenarios so that not only can you observe what goes right, but you can observe what goes wrong. And you can do this at scale and think about these edge cases, the recovery demonstrations and so on that will make these systems not just performant but resilient under failure and under unexpected conditions. So this is kind of a funny example, but this is something I was thinking about, which is, you know, if these systems really, if we get to the point, you know, this is a long path to get there.

34:18But if we get to the point where these systems really are able to provide that level of care that's like the best surgeon on the best day of their life every day at scale, it begs the question, you know, what do the human surgeons do, right? the people who've dedicated their career to this, one potential avenue for them is creating the, you know, the training data of mistakes that machines can learn on for forever, where, you know, you have people working in scenarios where they figure out what are all the failure cases of what, you know, how things could go wrong in surgery. And then we continue to, you know, use that to feed the machines.

34:55But what's your take on, you know, what is the role, What does the role of the human surgeon become once we get to this point? Yeah, totally. I mean, I think there are certain cases. So I personally don't think that human surgeons will go anywhere. And that's because I think there is, you know, the way how I see, and we haven't talked about this, so this is perhaps a good moment to think about this, is that already today we're unable to offer the care that really we probably should be able to offer, right? Like, I mean, if you have a – if somebody has a condition and they want to see a surgeon for some procedure to be done, the wait time is very, very long, right?

35:41And so because of that wait time, people might not be following up. And then if it's just something benign and it's really mostly for convenience that they were to get this procedure done, then maybe nothing bad happens. But there are people who are lost to follow-up simply because they cannot schedule the care that in an ideal world they need simply because the hospitals are at capacity. They just cannot offer sufficiently – like these appointments sufficiently soon. And there are other cases where it's not just about can we schedule people, it's emergency procedures that people might not be able to receive because the hospital that is closest to where the accident happened just didn't have the right expert in-house or the right expert there at the moment that this happened.

36:33And so they weren't able to receive life-saving care that would have gotten much, much better outcomes. Stroke is such an example, right, where not every hospital has the right specialty on staff. And so then people are evaluated and you determine that they need thrombectomy, so reperfusion of the brain. But it cannot be performed at the hospital where they are right now. And then they have to be transported to a hospital where they can perform this type of care, which delays treatment. and therefore deteriorate outcome usually because what they say in strokes, time is brain. A lot of people have heard that phrase.

37:17And so if you think about the ability of having a robotic system there in that stroke example, right, where the system has, where that hospital has access to an endovascular robot, that can perform the procedure regardless of whether they have a radiologist on or an interventional radiologist on staff. And right there in the moment, the outcome might be very, very different for those types of patients. And so I think that it is not just about are we replacing surgeons? I think it's the wrong question to ask. It is about can we offer the care to the people who need it in a sustainable version? And so what I think is most likely to happen is that some of the more routine care that right now is being caught by patients or by surgeons, what is being performed that they might not find particularly challenging, that some of those aspects of those surgeries, maybe not all of it, but some parts of it.

38:24For example, exposure where you make the initial cuts and you prepare everything for the safety critical parts of the surgery or the wound closure at the end, right? Like putting stitches and making sure that everything is fully sterile and completely ready for draping to finish the patient up. that some of those components can be performed with the system's help or by the system directly, which essentially just frees up bandwidth in order to treat more people, in order to provide more people with that same care. And I think that there are developments that will require us to rethink how care is being delivered simply because we do have an aging population which will require more care, which is fantastic, right?

39:10We're expanding the lifespan and health span of people, which is great, right? Like people live to older ages and they can do stuff healthily, which is great. But part of that also means that the reason they get there is because we're able to treat conditions. And so we need to continue to be able to do this, especially as a larger proportion is older and might require more care. And then there is another thing that is fantastic in principle but challenging if presented in this context, which is the fact that we develop new procedures and we can treat more conditions. So in fact, not just can we offer more of the same, we can offer more of different procedures.

39:55And so just in general, we can offer more. And how can we offer more in a surgeon-constrained bandwidth and capacity-constrained environment? This is not exclusively an do we do this autonomously or do we not do it autonomously type question. But I think it is clear to me at least that advances in technology and advances in engineering these health systems and the way how we provide care differently is going to play a humongous role in making sure that we can offer the best possible care and we can expand what is possible to be offered in the health systems. And I think in that space, I think technology and autonomy will play a role.

40:40Not the only one. Not the only one. It would be naive to assume that just introducing autonomy immediately solves. So it will not, right? Healthcare is very complicated. But it will play a role simply because we have a set amount of people. We only train I don't know how many surgeons, not enough. And then we make them work very, very hard, right? Like we strain them really, really quickly. And so I think being able to make their life better so that they can do what they care about more, which is giving patients the best possible outcomes, I think is a great opportunity for us to advance health and patient outcomes.

41:21Totally. It also makes me think everything that you just said about radiology. And I think that radiology is kind of considered the canary in the coal mine for AI automation in healthcare. And supposedly, you know, the AI models that are able to – is it radiology? Yes, yes, radiologists, yeah. It's the canary in the coal mine because, you know, all these models are getting so good at identifying, you know, cancer and all of these other things that you're looking for with the x-ray. But actually, you know, there's more radiologists than there ever has been because of the technology. You know, the technology has gotten so good, so more people can do it.

41:59Makes me wonder if a similar thing would happen with surgery, where, you know, if we perhaps, you know, not necessarily lower the barrier for entry that we expect of our surgeons. But, you know, the surgery process, you know, is augmented with this technology. perhaps then more people go into surgery because there's more job availability because everywhere, you know, every hospital now can have these robots that can do X amount of the work. And then you can actually have a surgeon on staff, you know, in these places that maybe wouldn't have them before that can do, you know, a lot of the surgeries that, you know, perhaps only certain experts could do.

42:33Maybe we have more, we can support more surgeons and, you know, actually solve that labor problem. Absolutely. Absolutely. I think because of healthcare and, you know, the world really, but these processes, I think they are so tightly interconnected that it is always very difficult to predict what turning, you know, one little knob a little bit, what the downstream effects might be. And so I think exactly as in radiology where it's – radiology is nice because every single image comes with a report. So essentially you have the sample, right? This is what I see and this is what the radiologist saw.

43:18And the images are on grids. So they're very, very accessible for a computerized system to read and process. And it happens to be a computer vision task where she learning tended to be the strongest from the outset, right, back from 2011. Very, very – it's really good. But I think a lot of things that have happened exactly as you say is that it increases demand because the first thing that you automate is not the fully autonomous read. It is workflow related. And so now I can, in fact, offer considerably more reads to patients and so on. And so that drives up the volume. Or I enhance the efficiency with which these scans can be taken because I have better reconstruction algorithms.

44:10I have a better way of – and so now I just increase the volume. And that suddenly increases the demand because I can now offer similar care. So I think a lot of these effects might be happening. And I would argue that in surgery, we'll see very similar tasks or similar effects as well, where I think that just because we will be able to treat more patients, the surgery adjacent proficions, whether it's going to be the surgeon or whether it's going to be a new type of job that is related to this is perhaps going to do exactly what you're saying, is like grow and create more opportunities for people to go into that field.

44:55But I think healthcare just generally, I think, has these very complicated cases where people – this is the problem with illness and disease, right, is that the biggest issue about that is that it is a diversion from a pattern. So, you know, and we call this machine learning. Previously, we called it pattern recognition, AI and machine learning, right? And then previously when there was not as much hype, people like to call it pattern recognition. And ultimately, not much has changed from that pattern recognition way of solving the problem. But exactly connecting this back is that disease and illness is a deviation of a pattern.

45:43And that makes it very hard to develop these types of algorithms robustly for something that is different than the norm because the algorithm by definition tries to learn the norm and not the deviation. And so there is how we offer the care then also is going to be affected by what these types of systems will be able to do really, really well and what they might not be able to do as well as we are hoping them to be able to do it. For example, decision-making under uncertainty might be something that we just don't want to give to a system because we want to have the human touch and the human decision-making and everything that is human, ethics and values and things like that.

46:31We want to have that decision-making process as part of it because as society, we feel better about these decisions being made by humans. There's research on these types of things, what people tend to be okay with being decided by an algorithm versus where they would want human judgment in case. And they tend to be forgiving even if the human is wrong because some decisions, they're just hard and so it's very difficult to automate them. So this is where I think it will be close to – I don't see humans being taken out of the care delivery simply because when people seek care, they're vulnerable.

47:15The connection to people who care is just as important as the actual procedure that is being delivered. And so I don't see that going anywhere. On the other hand, as we discussed earlier, people have variability. Some days, you feel like you can only win. We have days like that and then there are other days where you feel like this is just not – today, it's not it. And the problem is there when you perform surgery, you not having the best possible day means somebody else not receiving the surgery that they could have received the next day. And so they're having these assistive systems that just pick you up and make sure that you having a bad day doesn't result in a patient having a terrible one is really a first grade way of just making sure that you're not taking out the humans.

48:14You're just making sure that they have complementary awareness. They – you try to support them, deliver the best possible care, which they want because, again, surgeons and clinicians in general, ubiquitously, the one thing that they care about the most is making sure that they do the best for their patients. And so, you know, when you offer them the ability to do exactly that, they will not be technology adverse. they will be critical as they should be because they're not playing a game. They're trying to treat people. So they will have hard questions and you better have good answers. But once you can provide the corresponding databases and evidence that truly the incorporation of technology like that helps them treat their patients better, the clinicians will be the first one to advocate for this type of technology to be used routinely because that's what they care about, doing better for their patients.

49:19So this makes me wonder, you've described, and I really agree with and appreciate the vision that you've painted here as autonomy as like a ladder where you're slowly going up the rung, you know, one rung at a time and you're, you know, finding the systems where you can actually introduce that automation and where that does streamline things and then you slowly but slowly but surely you know go up the ladder and build up to you know potentially a fully autonomous surgery um what is the first what are the first use cases that you're targeting at inner logic and you know do you plan to build any of your own you know ai models for that will you use other models and and yeah what's what's your first target goal here that you're trying to go after with inner logic So I think we're seeing the full spectrum of where autonomy at the moment is the most mature and most likely to convert soonest for some of these applications.

50:18And I think the ones that we've been seeing, I think, are the ones where robotic systems are already quite mature and ready to take the next step in the level of autonomy that they provide. And then in the area where not acting is worse than acting with a little bit of a mistake. So to make this a little bit less opaque, I think the first area is musculoskeletal types of robotics where we're dealing with hard – like rigid structures like bones where we have much better predictability of how the bones will move. The instruments that are being used there also are usually rigid and relatively bulky.

51:03And so tracking them and understanding where they are and making the right decision is relatively easy because the perception problem for the system is just relatively easy. So that's one area. Musculoskeletal applications and musculoskeletal devices and procedures is one of the areas. The second one that is important for us is those emergency type of procedures where unlocking care in those environments is really important. So there we're focusing a lot on endovascular type of applications because of the stroke use case. There is a large interest also right now in endovascular robotics that are being developed for these types of use cases.

51:46And so enabling that type of treatment is important. Similarly to disaster response where you can think about conflict or other type of trauma that might happen where people are hemodynamically unstable, so bleeding. And being able to treat conditions like that to stabilize patients and make them ready for transport if an expert surgeon to do that is unavailable is another one of those capabilities that are potentially quite enabling. So these are, I think, some of the use cases that for the demonstrations where we want to showcase some of the tooling that we're building, that we see interest and that we see opportunities.

52:29But again, we're building computational infrastructure that is designed to help OEMs and leaders in that space solve their problems that they have in procedural devices and procedural autonomy at software speed. And so it is all designed to adapt to the use cases that our OEM partners seek to solve using the platform that we are. That's awesome. I'm very excited. I can't wait to see, you know, how you apply the software tooling that you're building and, you know, how you can scale these systems because I definitely resonate with your vision. I think, you know, the more health systems and, you know, individual hospitals and doctor's offices where we can get this capability, I think it's all for the better because you can give that care and you can equalize it and get it out to as many people as possible.

53:27I think it's very important. Better software helps us do that. For listeners who want to follow Interlogic's work, where should they go and what should they expect to hear from the company next? Yeah, so I think we will be releasing some of our initial technical reports and other types of information about the company soon. I think we've been not quite operating in stealth until now, but we also haven't formally announced. And so now there is the big conference for the Society of Robotics Surgery happening soon where we will be present and presenting some of our initial developments. So that's going to be exciting.

54:10So on the web and on LinkedIn is going to be the best outlet to follow us. I will say on the idea of autonomy being a ladder and working our way up, you know, you brought up self-driving earlier, which is a good example. The one issue with self-driving, which I think has nice parallels to surgery, is that driving a car is a very dangerous activity, actually. And there was a lot of examples of Tesla when Tesla was on the road and testing things originally where there was a lot of accidents and it got a lot of bad press. But Waymo was sort of like a leapfrog product where it came out of the gate.

54:50And yes, of course, they had testing, but, you know, comes out and it pretty much works at this point. But there's still times when even Waymo, which I think is probably the best self-driving product on the road today, you know, Waymos will still get stuck somewhere. And, you know, people, you know, take the meme of it and, you know, film it and everything. And I worry, you know, what, you know, obviously when you hear autonomous surgery, you worry about that use case and you really don't want your autonomous robot to get stuck mid-surgery. And then, you know, you have to have a mechanic on staff or, you know, somebody who understands the software to be able to actually like course correct it.

55:26What do you do in that sort of situation? Is that a reasonable parallel and how are you thinking about it? Yeah, it's a great analogy. And I think people have a very clear perspective on autonomous driving because it's been around for so long. And so we've experienced it in certain ways. And I think, first of all, I will say I personally think Waymo is a great product. I use it a lot. And even my kids are convinced that it's fun. But, of course, the technology can fail. And then it usually makes the rounds and people like to joke about it. What I personally think, however, is that that reaction is perhaps not the best one because it's true, right?

56:13Like if we deploy these types of systems at scale, they sometimes just will get it wrong and they will make a mistake. And then, of course, we can run our mouths and we can laugh about it. We can, you know, be cynical, which as a German, it's close to my heart. I understand. But it somehow misses the bigger picture. And the bigger picture, I think, is that when you look at the data, and there was this fantastic op-ed in the New York Times late last year that was describing autonomous driving as one of the most important public health interventions in the recent past because of how much safer it is compared to the data of real-world drivers, right?

56:58And that's because, yes, the system still fails, but hey, normal drivers, human drivers, they fail a lot all the time also. And so I think this laser focus on individual mistakes, I think, is just getting the big picture of the story wrong. And the reason I want to bring this up is because as we are starting to introduce this type of technology in surgery, which admittedly is a very high risk domain and we need to make sure that everything is done up to standard and validated and done safely, of course. But just because of how we cannot avoid mistakes, medical mistakes in regular practice, because it's hard, we will very likely not be able to avoid individual mistakes sometimes either.

57:51And so this is where I see some of these parallels where we will have to do a really good job in communicating and convincing that this type of technology ultimately at a population scale is going to offer tremendous benefits. Now, again, we're not there today, and it will take us time both in terms of building the systems to that level of standard where they can perform at that level and then demonstrating that truly they are safe, which will not be tomorrow. It will take a while. But I think there is going to be very likely a similar type of story where we will appreciate the introduction of these assistance systems and autonomy subtask features as a similarly transformative public health intervention in order to ensure better patient outcomes.

58:44Do you ever think it will be seen as unethical to let a human perform a surgery versus an autonomous surgeon when we get to that point? This is not an area that I can speak with authority. But I think that already today we see that there are certain tasks in surgery that are very mechanical and they feel repetitive. So, for example, if you have to perform different sutures for anesthesiosis where you join two tubular ends together, We know that if you can automate this robotically, the repetitiveness and the reproducibility of the different spacing between the stitches, the robot can just do it much better than the human because, you know, it is robotic.

59:36So it does it with a much higher level of precision, which creates a much more laminar flow profile, enhances the overall pressure that can be applied. And so it's just better. There are proof points like this. And so it's quite possible that truly once this is validated as safe, as a real product level and ready for deployment, that we would feel silly not doing it the new way because it's just demonstrated to be better than what we can do otherwise. And maybe that threshold to meet would be lower than like the self-driving threshold because not everyone is a surgeon, right? So not everyone has a certain level of expectation or self-confidence in their own ability like they do with driving.

1:00:25Whereas I think driving has a much higher bar to cross for the regular average person because they're like, well, I drive and I think I'm a safe driver, whether or not they are. So they have a certain level of like, well, this has to be above and beyond safer before I accept it. Whereas surgery, perhaps the bar is maybe lower for the general person because they don't do surgery. So they don't have any sort of threshold with which has to be crossed before they'll trust that. I guess obviously you need to know they can do it without cutting you open and all this stuff first. But yeah, cutting you in the way they're not supposed to, I mean.

1:01:05I mean, surgery is a highly regulated field. And so I think once you really make it through the different regulatory bodies and into a product, There is a good body of evidence that the system is up to spec and if used correctly, will perform as indicated. Now, this being said, I think there is a lot of conversation to be had in order to build trust about this type of technology. Because while the average person is not a surgeon and therefore might not have the same level of appreciation of what makes or doesn't make safety, I think people are also quite concerned about their health. and they don't want this concept of trust on what care they are being provided with, I think is incredibly important.

1:01:54And whether we're ready to accept a machine performing part of this, I think it will require a little bit of conversation. However, if you look at elective procedures where people don't need the surgery now, they can choose where they receive it, we already see that patients vote with their feet and go to centers where robotic surgery for the procedures that they need is being offered because of those demonstrated benefits of, you know, reduced tissue trauma, shorter recovery times, and so on. And so this is already an effect that we see. It happens. And doing that goes to trust. Yeah, that's awesome.

1:02:36I didn't know about that. So, you know, obviously we follow at the Neuron pretty closely, you know, the developments of large language models and how they're progressing. And now we're really judging, not necessarily based on the model itself and how it responds, but how it works inside an agentic system. And robots, famously, are a very difficult problem. And a lot of people say, in order to solve robotics, you need to solve robotics. What would you say is the hardest part about engineering for robots, especially surgery robots in this case, and how are you trying to approach that and do it differently?

1:03:15I think by far the biggest issue is that you have, right now you have robotic systems that are designed the way that they are designed. We have this big benefit in surgery. We have this big benefit that, in fact, the data flywheel exists. So we already have surgeons that are using these robots to perform surgery every day. And so this demonstration data exists. That's one of the big benefits that we have in surgery compared to, say, general purpose robotics where, you know, data on how to fold a T-shirt using a robotic system, it just doesn't exist. And so if I want to automate that with a humanoid robot, my first step is to go and acquire I don't know how many demonstrations of folding T-shirts, which is not a particularly pleasant thing to do.

1:04:10But it's important because that's what unlocks the task. Now, in surgery, again, we don't have this problem because we already have surgeons acquiring this type of data. But the visibility of structures, the knowledge required in order to do surgery safely, the delicacy with which the tissue needs to be manipulated precisely, the viewpoints that have to be selected in order to do it safely. I think all of these are incredibly hard challenges that while we're seeing that in a very narrow context, we can automate them. So we've seen clipping and cutting, for example, of the bile duct fully autonomously in live animals.

1:04:59So while we're getting closer, doing this robustly, especially across potentially diseased and therefore very inaccessible anatomy, is incredibly hard. And so this is where I think like we will first see some of those more predictable type of procedures that are being automated and those assistive tasks don't even require necessarily tissue manipulation like camera control and things like this, that they will happen first. But yeah, I mean, exactly as you say, in order to solve surgical robotics, you have to solve surgical robotics. And that is about as hard. I think with what we're building at InnoLogic, our hope is to take a lot of it from real-world experimentation, which by design is slow because you have to build a mature prototype.

1:05:53You have to do things like you can acquire with a real physical prototype. You acquire data at 60 minutes an hour, right? Like it doesn't go faster unless you build a same prototype. Now, we're trying to take a lot of this. Because the real data will not go anywhere because, again, if you want to learn what the best surgeon does, you're going to learn from what that best surgeon did, this real data. But for a lot of it, like the demonstrations about recovery, padding it with having the system understand robustness and so on better, a lot of that can be taken into software. And that allows us to speed up some of this development and especially speed up iteration cycles.

1:06:36And so this is one part of what we're building in a logic in order to make the development of these types of algorithms and systems that power this push towards autonomy and better procedural devices to make it less painful and allow us to iterate faster. is there a role for uh like sensor data in that like in the same way that waymo you know uses a lot of different um i forget exactly what they're called but they use a lot of different sensors beyond just cameras is there you know to any of the robots that exist today did they collect multiple different types of sensor data when they're doing these surgeries could that be useful to the process i know that's a lot more data than you have to process on the software side No, absolutely.

1:07:25In fact, one of the big challenges when surgeons went from conventional laparoscopic surgery where you take these long scissor-like instruments with long shafts and you manipulate like this for minimal invasive surgery, when they went from those types of procedures to robotic, one of the biggest issues was that one of the sensors that your hands had built in suddenly was gone. And that was force feedback because the robot does not communicate to how strong of a force you are applying. And so you have to infer the strain that is being applied to the tissue from the visual system. So you have visual haptics.

1:08:11How hard are you pulling based on the deformation that you observe with your eyes? That is something that has to be learned. And so we train surgeons to do it this way. But new systems now bring, for example, these four sensors back into the device. And so you know quite precisely how strong you're pulling and you can incorporate that into decision making. But as I said earlier, right, like this differentiation that individual manufacturers have for their devices and how they want to use it is different. So we're accommodating it all depending on the specific use case and partner. Cool. Very cool.

1:08:50Well, I wish you all the luck with that because I think, you know, a general surgery robot system would be really amazing and could benefit everyone, you know, especially because if you're learning from every system and learn how to work with every system, I think that would be the best case scenario because then you're building the best system possible that everyone can benefit from. So really appreciate what you're doing. Really appreciate you taking the time. And Matthias, thank you for joining us today. Yeah, thank you so much. This was fantastic. Real quick, I want to take a second to offer our thanks again to Dell Technologies and NVIDIA for sponsoring today's episode.

1:09:26For our full content hub on AI-native factories, hybrid AI workloads, data readiness, and sovereign AI, head over to the Enterprise Guide to Scalable AI Hub on techrepublic.com or click the link in the description of this video for more. If you liked today's video, please take a moment to like and subscribe. And don't forget to check out the Neuron's other projects, including our daily newsletter, which is read by more than 700 ,000 people just like you. The Neuron Academy, where you can learn all about AI and how to use it in your work and life. And also check out our new sister newsletter, if this topic is interesting to you, Robotics Insider.

1:09:59Thanks for joining us, and we hope to have you back next time. Farewell for now, humans.