Catherine Mohr 0:07
Thank you so much for that kind introduction. And I'm really honored to be here. I work in the world of VC, as well as in the world of long term product development. And so while we're thinking in the VC world, when's my next round going to be raised? When am I going to get that next approval? How are we thinking about, you know, I'm going to bring out this clinical trial, a lot of the things that I work on, have much longer timeframes. And so I'm going to be talking about the context of some of the things that we do in terms of trying to solve some of the wicked problems that a lot of these companies are designed to do. But in a longer context, and a longer timeframe. Because a lot of the problems that we're looking at when we're looking at this intersection of healthcare and technology, are fundamentally about trying to address issues in patient care. These are the important problems that we are facing. But a lot of those issues stay consistent and have been consistent for a very long time. So when Intuitive Surgical, was put together, so the first prototypes were 96, the company was 99. FDA approval was 2001. It was trying to solve the wicked problem of how do you make minimally invasive surgery, which is better for the patient, the incision serves no therapeutic purpose. If you can make that incision as small as possible and do exactly the same surgery, it will always be better for the patient. So they were saying, Okay, this is the wicked problem I need to solve. Can I do it with robotics. And so you can see here, some of the early prototypes that were coming out of Sri, trying to figure out not just sort of surgery at a distance, but that distance being across the abdominal wall. And early prototypes, like most early prototypes were kind of ugly, in prototype be looking. And when you think about being able to see what this could do. That's the point that you find yourselves at so often, in the early startup phase, when you're saying I'm trying to solve this problem, I have this technology, the technology is too ugly right now. And it's my job to get that technology to the point where it can solve this beautiful problem. And so, after a few iterations, you get to elegant solutions to things. So, but you all have to take a technology from that ugly phase into that usable phase. And so everybody is familiar with this. And, you know, we heard in the panel before, often people are like, Oh, will you try out my technology, those are the early adopters, those are the people that will move out forward. But getting everybody else to jump on board is where so many companies die. And we are in a hype cycle right now, around AI, we have been in hype cycles around cloud, we have been in hype cycles, we wet we, you know, if you are focusing on the real problems, you will weather through these hype cycles, and you will get to cross that chasm. So we think about robotics, as futuristic, we think about it is really modern. This is a 29 year old overnight success. And so when I'm going to be going back and forth in this timeline a little bit, because where Intuitive Surgical is now was not obvious, or even predictable from all of the stages that it was in along the way. And there were plenty of things that could have killed the company over and over again. But when you look at that takeoff in 2019, in general surgery, that was an introduction of a new product that was specifically tailored to general surgery, because people had not seen enough uptake in that area. And so it's about iteration, it's about understanding what was great about the product that you had, and then taking it to that next step. And so today, like I said, it feels like it was inevitable. You know, today, almost every 17 less than 17 seconds someone has a DaVinci surgery. And and so it feels like a really dominant position. But early on, it really wasn't that apparent So I went to medical school in 2001. And that's when we moved up to the San Francisco Bay Area. And my husband joined this little startup company called Intuitive Surgical. So it was safe to say that there were almost no procedures being done. At that point, they were in the order of hundreds per year 2003, there was a reverse stock split. Intuitive was struggling to keep its stock price over $10. Just doesn't seem like that's possible from where it is now. But understanding where we were in those days, is we were building something that a lot of people didn't understand, and didn't see the future value of. So I joined in 2006. And maybe a total of 100,000 procedures had been done. At that point, everybody was still not so sure this was really a thing. So Medtronic secretly started working on its own robot in 2013. They said they'd been working on it for three years in 2016. So that's sort of the DLR days when they were exploring this j&j. And Google got together and created a joint venture in order to make a robot and both of them in 2015, the robot was going to be out in about two years, they were sure. And of course, we all hit the global pause button in 2020. So we didn't have a whole lot of growth. But if you're looking at where you can see the idea, and where you can appreciate the idea and where other companies say, oh, maybe we ought to do this, this is it's not inevitable, as you're going along. And so these are very important things to think about, as you're thinking where you are in any one particular point. But this is why I joined Intuitive Surgical, not all the rest of the robot, that instrument, because that was the big idea that was solving the big problem. The rest of the robot is just to bring those instruments into the patient through a very small incision. And it allows you to work inside the heart without cracking open the chest, but it allows you to take therapy, surgical therapy, and divorce it from the incision, which, as I said before, serves no therapeutic purpose. And so intuitive has, over the years, evolved with different versions of being able to bring these small incisions in to being able to bring a cluster of instruments in through narrow access, and be able to expand inside at a distance from where you place the incision. This allows you to do things that we weren't able to do with triangulating in through the ports, it was fundamentally different, and a fundamentally different approach to the same problem. And this allows changes in narrow access surgery, being able to go trans orally, being able to if you're going to bring a colostomy to an extraction site, and overall be able to move it down to a single port. Not an an obvious continuation of the same procedure. And so by allowing these different kinds of access, you're putting different tools in the hands of the surgeons who will teach you about how to use your products. And overall, the common goal has to be improving outcomes. And so there are the robots out there. But fundamentally, none of these robots, Intuitive Surgical or any of these, we're not competing against one another. It's really about the status quo. It's really about how is it being done now. So when you think about your competition, it's generally not other companies that are trying to enter the space. It's the established procedures of how that therapy, you know how that particular pathology is treated today. And so it's it's a different take on competition, but your competition is status quo. In the process, I ran research for about nine years, and then I went on to run
Catherine Mohr 9:35
the strategy in research, we were always focusing on the wicked problems. What are the problems that we try to treat? Where is the gap between what surgeons want to be able to do for their patients and what they can do for their patients? And can we fill that gap with technology? So it's not I have a robot? What do I apply it to? It's I have a really big clinical problem, the gap which I can close with robotic technology, what robot should I make to close that gap? And so in the process of being in running research, and then later strategy, I wrote business plans for essentially every robot that is out there in the world right now plus a few more that I might tell you about another time. And they were all questions around what is the big problem that we're trying to solve, and we didn't build any of these robots. But then a new technology came along. This was a shape sensor. And it had a fiber Bragg grating that allowed us to look at the pose at the tip of a meter long fiber in very rapidly, and with very, very high precision. We already had our problem list, we already knew what we were trying to solve. And so when this technology came out, we knew what we were going to do with it. And we built a new robot around it. It's not like we came with the technology and said, What am I going to apply it to? It's I have this laundry list of problems. And new technology has come out which of these problems does it help me solve. And so around that shape sensor, we created a catheter robot. And this is just like that wrist that I talked about. That's the thing you believe in the tip of the catheter is the thing that I believe in out of this, we built the robot around it to put the tip of the catheter in the right spot. And so what was the problem we were trying to solve sort of a surgical problem, but also a little bit of a diagnostic problem, the big gap in lung cancer diagnosis. So the National Lung Cancer Screening Trial was going on in 2010. We were in keep in mind, so you're working with data from the time to try to predict where everything is going to go. And back in 2010, the National Lung Cancer Screening Trial was trying to do something that no trial had ever shown before, that it was actually cost effective and reasonable to do lung cancer screening. Because every other trial in the past had shown that you harmed more patients than you helped if you did mass lung cancer screening. Because the workup associated with the the the nodule that you saw, was so damaging that you actually hurt more people than you helped. In the early stages of nodules, nine out of 10 of them are entirely benign. And it can be very difficult to differentiate, which is cancer and which is not. And so lots and lots of people were getting surgeries, for nodules that didn't have cancer in them about 30% At that time, and people were waiting on this treadmill, see a nodule. And then you do serial CT scans, and the patient believes they have cancer for 18 months. And don't underestimate the burden and the psychological burden of the workup. This is not just cost. This is real patients suffering associated with this. If the nodule has actually grown enough, often you've halved their life expectancy. In the process of waiting for that nodule to grow big enough for you to determine that it was actually cancer. You do have that procedure, you do a lobectomy, and we're really talking about an 18 month cycle before resolution. What we saw with that catheter was a completely different paths. There were bronchoscopes out there, there was even a robotic catheter that other people were looking at. But without the accuracy to get someone off of that diagnostic treadmill, it was not going to change that path for somebody. But what we could see with that piece of technology was a four week cycle of a nodule identified being able to snake in and be able to no matter where it was in the lung. And no matter how small it was, at the you know, centimeter size, be able to get a definitive sample, give them a diagnosis right away and then be able to go to definitive treatment. And so what we saw in that piece of technology was the potential for a factor of 10 improvement on people on the lung cancer cycle. And so that was a wicked problem. And it was a wicked problem that we had, and we had looked at for a while and we had been waiting until the technique algae showed up that let us get there. And so this is the ion now. And we did our so in, I talked about the catheter being everything, or the end effector being everything. The other problems that we had to solve in this are, we had to be able to do auto segmentation of the lungs, we had to be able to do pathfinding, you have a synthetic model of the lungs. Over on the left, you have your actual visualization. And you'll notice, the path tells you to go in a place that's completely obscured right now, you wouldn't have seen that branch on the airway, you need the synthetic model and the real time view to be able to navigate your way through the lung, you'll notice this is retroflex. This is down the airway all the way back up into the apical segment, being able to get a full size sample that is definitive and able to give that person a diagnosis, we had to be able to get there in order for this robot to do what it was that we needed. And so we were waiting until the technology was there. So, and I promised one funny story about the development of this, which is you can see the user interface over there. Over when when you're building these new things, you are the first to try to solve some of these problems. And we went down to Fry's, when it was still open in Sunnyvale, and bought two track balls, because I was like, oh, we need to be able to think in terms of pitch and yaw and in terms of rotation. And so we mocked it up. After a trip to fries with two track balls, there was a team assigned to really redesign the entire UI and make it super Excellent. And all of this sort of thing. After nine months of really trying to get something better than to track balls, we have to track balls. Because it, we were solving the problem in like, well, but we need this kind of positional awareness. And we need this kind of positional awareness, and those are separated. And so that should be the solution. And sure enough, so there's really random stories. And there's a few people out here in the audience who have been along the way, where if you knew the origin of a particular thing, especially in medicine, where it's so concerned, you know, everyone is very specific. And we have all of these processes, how the occasional just, oh, I have an idea to trackballs ends up in a longterm product, I kind of there's little funny stories like that all over these robots. Um, I couldn't give you a talk about the future without talking about data and AI, because that would not be, you know, that I'd be shirking my duty in the hype cycle to, to not address this. You have to be thinking about data and AI, from the point of view of what problem is it solving? What important problem is it solving? Not? What nail am I going to hit with this hammer that I have? And so when we think about the data that are available, we need to be thinking about what do we do with those data, it's not just enough to have it all we need to be able to interpret we need to have a real reason that we are collecting that data. Now, one of the things that collecting a lot of data lets you do is to drive sort of overall insights around you know, from from mass sets of data about how people are using it, what are you know, what does well for our workflow? How do you think about pulling all of those sorts of things together. So that's the kind of operational type data
Catherine Mohr 19:04
we also bring things on to the robot in terms of helping to solve clinical decision making type problems. But these are not necessarily all data. One thing that we're using machine learning and image processing for is, as I mentioned, in finding those airways, in for the ion, this is being able to take a preoperative CT scan or an MRI of a patient, being able to build 3d models being able to do path planning. But, you know, in most cases, you don't actually need something like this, or it doesn't necessarily change. The practice doesn't, you know, in medicine, we say don't measure, you know, don't do a test that you're not going to act on the results of that test. And so don't bring technology in if it's not going to change any of the things that you're going to to do. So, in the cases where preoperative planning really does change what you're going to do, it's an excellent technology. But it's not necessary. In most cases, if you can see everything we did bring fluorescence imaging on, this was the wicked problem of, I can't see if I've left a positive margin behind, or I have trouble identifying this particular structure. And we're like, Okay, well, what if we make glow? Oh, hey, that would solve that problem. And so we brought fluorescence imaging on way before there were any imaging agents to highlight all of the things and we created a platform, I actually stood up at a optics conference and said, We're going to bring this out, if you guys want to make imaging agents, this is the wavelength you need to fluoresce in. And it spawned a whole bunch of different companies, because they were going to have a installed base to be able to bring out their product into so this was thinking in terms of, do we want to be a drug company? Do we want to, you know, or do we want to change the way imaging is being done and give a lot of other small companies an opportunity. This is what I am most excited about using data for. And that is training. That is understanding how people go from being a novice to being an expert, how we think about the way lots and lots of experts do something so that we can look think about best practices. This is where I think observing and what somebody is doing, and being able to give them real time feedback about their performance gets into that virtuous cycle of people being able to get better and better at things. And so this is taking some of the data that's on the robots, and taking it off line, taking it off the robots and really starting to think about things like education and training. And so this is the real application that I'm the most excited about in terms of things that we can do with AI. And we've been able to look at the way surgeons move the way they, you know, optimizing paths, being able to think about ways in which we can differentiate between experts and novices and be able to bring people in that into that expert cycle. We can look at, how does a, you know, how long does it take a novice to do, we can deconstruct into the various tasks. And then we can say, what does an expert look like doing those same tasks? And how do we think about transitions, and what is what's the guidance that we're going to give, you can also do things like medicine, as an apprenticeship model is very dependent on Oh, I have all of these cases that come in, and I do these cases. And eventually, I will do all the hard edge cases of things, I will see all of the difficult stuff. And I will train on that. But you can distill all of the difficult stuff, and drill people on things that they might see only once every 1015 or 100 cases, and get them to excellence on things they don't otherwise have the time or the case volume to be able to always see. So that the first time they see that very difficult edge case, they are already practiced at it. And they're already good at it. And so this is not education by exhaustion, where we just have it over and over and over again. It is intentionally taking what it is that we are being able to glean from this and say, what ought we to do in terms of practicing and drills and, and really moving people towards excellence. And so being able to get this kind of feedback. And this real time kind of feedback, I think is the place where we're going to, you know, where I am going to hit things with that technology hammer, because it is a wicked problem, having enough people to be able to give real time feedback so that people can continue to get x more and more. Excellent. So now I'm going to jump around in the time and load lines a little bit. This is today. This is well this is 2023 the number of robots that are around the world. It looks really big. When I started doing strategy when I started as VP of strategy moving from research into global strategy, this is more what intuitives installed base looked like. So Asia there were few You can see a few cases, there are few robots there. And we were really trying to understand what was going to drive adoption of a lot of these technologies in a lot of other areas. And so this is now where I really go back in time. I take a very long timeframe, look, and I went back and I started looking at these economies. And I started looking at where are they on this trajectory? And how are we thinking about this trajectory overall. So I'm going back 100 years now. This is life expectancy, and income per person today, normalized dollars each.is, a country, the size is the population size, and the color coded for regions. So you can see Europe, you can see the blues are Africa and India, India and China are still really big. And you can draw a line through the middle because we like to draw lines through the middle of our data. But this line is sort of the health wealth line. This is how much healthier an economy gets when it gets wealthier. So public health, better nutrition, education, all of these different kinds of things. But you'll also notice that 100 years ago, pretty much no countries were over 60 year life expectancy. And that's not the people dropped dead at 60, or 35, as you can see in India, and China is the child mortality was off the charts. So we had bifurcated societies, we lost lots of children in that first one to five years. So let's play this forward. Now, World War Two breaks out and everything goes kind of crazy. something amazing happens after World War Two. You'll notice that even though with today's normalized dollars, a lot of the poorest economies are still saying poor. Everything moved over that 60 year life expectancy. And that health wealth line flattened out. If you were very sharp eyed, you would have noticed that those big Asian economies went up in health. And then they started to move along the wealth line. So they got healthier, and then they got wealthier, because they now had the young members of their economies driving this. So what happened after World War Two vaccinations, antibiotics, and really big changes in surgery and surgical availability. And people might go surgery, really should surgery be up there with those pharma and with vaccination? I will absolutely say yes, because we do about 300 million surgeries a year around the globe, we're missing about 150 million people who needed a surgery who aren't getting it. That's resulting in about seven to 17 million avoidable deaths if somebody just had access to surgery where they were. And to put this in perspective, we have 3 million deaths a year of HIV, TB and Malaria combined.
Catherine Mohr 28:22
Globally, and we have 10 million deaths a year globally from cancer. And so if you don't think that lack of access to surgery is a global health crisis, then you're not looking at the same data that I'm looking at. And so when you say how are we going to fix this? It was these realizations as VP of strategy, where I said we need to solve a much bigger problem, then our robots the right thing for these different economies, we need to start really looking at the wicked problem of training this next generation of healthcare providers to meet those really critical needs. And when you think about infrastructure, you tend to think about, oh, I need to build hospitals in emerging markets. No, you need to train people. Because it takes a year to build a hospital, it takes 10 years to train the people. And when we're thinking about the length of time that these economies are changing in. It's really, really critical. Especially in areas like Ethiopia here or Sudan, where I'm posing with all these medical students. The bullets started flying about three days after I was there and I flew out, and all of their professors have fled the country since then. And we're still training them remotely because there is no other option or we lose an entire generation of the next Sudanese medical professionals. But medical training, and I'm running a little late I'm sorry, but I will try to speed up a bit. But I love this stuff. medical training is a wicked problem. These are the same economies over the same period of time. And this is an exponential adoption of technology. This is cell phones, difficult to make where you make them and easy to deploy, where you deploy them. This is the technology, slope, we hope for all of our technologies, as we are building things. This is what it looks like trying to do training. Because we have an apprenticeship model. And it's not that the same economies over the same period of time wanted doctors any less than they wanted cell phones, it's that there is a fundamentally different mechanism at play for the way we train versus the way we can deploy technology. And so we built our training materials for replacement economies. So when you look at the United States, and you look at Europe, this is now age zero to 100. And age distribution, Americas Europe, Africa, Asia, this is 2020. And this gives you an idea of, you know, the inverted pyramid that's happening in Asia, that sort of curve came in with the one child policy. But we built training programs for economies that looked like Europe and the United States. We built them for ones where it's a guild system, and you're replacing clinicians with new clinicians that are coming in. And it's working its way through, we did not build them, for the economies of Asia and Africa that are growing. And we move this forward the next 70 years. And you look at the economic development models, you look at the best modeling we've got on birth rates, and you don't have a replacement economy, you also are looking in the wrong place for your future customers, if you think all of your customers are going to be in the United States and Europe. But we don't have a training program that works with this. So when you think about the the modeling who your customers are going to be in the future, and what you are thinking about and what technologies you're bringing forward. Understanding that what has been before, all of the things we have been thinking about since World War Two, that this world order is changing dramatically. And several of the economies that are going to be in this top 10 Over the next 30 to 60 years, are in these rapid growth areas where we don't have enough way to train where we are where our training materials are falling woefully short. So scaling surgery, yes needs hospitals and equipment and practitioners and midwives and medical officers and all kinds of things that we think about in the global global surgery world. But scaling it exponentially needs a really different take on education. And this is where this is the wicked problem that I want to bring AI and ML into. Because we can scale reading based knowledge. We can scale lecture based knowledge. If you can learn from a lecture, you can learn anywhere, but we haven't scaled skill based knowledge, we haven't scaled, can I do something be observed by somebody and then be able to move it forward? And so assessment is the key to this kind of skill training. We right now do so see one do one teach one do one under observation. We need to shift it to see one teach yourself one with skill, observation and guidance, and then do one to be able to fully scale this. And so I left strategy and put on a foundation hat. And we ran the surgical education learners forum bringing together understanding how are we going to really change technology of training, we have partners from all over the world. I'm going to skip through a few of these, the winners were out of Ethiopia, and they are using a peer to peer video, AI review. And that allows learners turns learners from a burden into an asset where they are reviewing one another's videos and being able to give formative feedback. And we had whole hospitals in Cameroon where everybody from the trainees all the way to the senior attendings all learned laparoscopy together without any instructors coming to the hospital and they are now We're doing nine to nine to 12 laparoscopic cases per week in that hospital. So when we look at being able to pull all of this together and think about trying to solve these wicked problems, a lot of the work we're doing is looking at novel ways in which people are delivering content. Looking at, as I said, that video based and AI based peer review for skill review. And we are building recommendation engines, where somebody says, Oh, I need to learn how to do this procedure, we have trained a recommendation engine using all of this reference material, and we can offer them the best match for their training needs to be able to get them in the environment in which they're in to the skill level that they need, and then be able to get feedback from that. So we're building these as an ecosystem, as a community of practice. And we've got partners from all over the world. And what's exciting this year is both the College of Surgeons of East Central and Southern Africa, and the West African College of Surgeons have formally adopted these training methods into their training programs. And this next generation of surgeons coming out of Africa will be using these. So we are investing in these kinds of technologies to fit into this larger ecosystem to solve this wicked problem. And there's a lot of space in this world. So pulling it back. When you're looking at, where am I going to apply technologies, what problems Am I trying to solve? There are so many, there are so many different places where all of these problems are. And there's so many niches in which your companies can be in to try to solve these big problems. And so it's it's sort of yes, you look at something like Intuitive Surgical, and it's big. But I jumped out of making robots and said, I want to redo how we do global training, because it's the focus on those kinds of wicked problems. And we've got companies here that you're going to see all around who are solving this. And you know, the one on the left is the visit plate from a VC, which is solving a wicked problem associated with glaucoma surgery, where and what they found was an anti fouling coating that allowed them to be able to move forward with something that I think is going to revolutionize. But there's also robots that can be out there. This is spearos intubation robot, and is just being able to steer and visualize for very difficult intubations. And so there are so many niches out there, when we're thinking about bringing in new tools for changing the practice of surgery. And then looking at that long term goal that we've all got a fully digitally in integrated and being able to do everything that we want to do with the data. And the path along the way, is really focusing on these wicked problems. So with that, sorry, went so far over, but
Catherine Mohr 38:17
we're all back at this intersection of healthcare and technology, and really keep your eyes on what is it that we are doing for the patients? What are the wicked problems that we're trying to solve, and then bring on bringing the technology to do that? Thank you
© 2024 Life Science Intelligence, Inc., All Rights Reserved.
