edd9164d216c19945bea55d0825befe1a07fdae5.jpeg

Welcome to my podcast. I am Doctor Warrick Bishop, and I want to help you to live as well as possible for as long as possible. I’m a practising cardiologist, best-selling author, keynote speaker, and the creator of The Healthy Heart Network. I have over 20 years as a specialist cardiologist and a private practice of over 10,000 patients.

Podcast Summary

Introduction

Dr. Warrick Bishop, a cardiologist, author, and CEO of the Healthy Heart Network, hosts this episode featuring Ryan Talbot, president and founder of ViewMotion, a company specializing in AI-driven movement analysis technology. The two met serendipitously at an airport in Santiago and discovered a shared interest in applying advanced computer vision and machine learning to health and sports performance, leading to this discussion about how movement analysis could revolutionize injury prevention and athletic performance.

Key Takeaways:

  • ViewMotion uses computer vision and machine learning to analyze human movement from standard smartphone video (4K, 60fps), converting video data into detailed movement metrics with centimeter-level accuracy by calibrating the scene with five cones.

  • The technology generates comprehensive reports including split times, maximum speed, asymmetries between left and right legs, ground contact time, flight time, and a step quality score compared against normative datasets based on age, gender, sport, and anthropometric values.

  • Coaches receive three prioritized recommendations for athlete improvement based on algorithmic analysis that identifies force deficiencies, velocity deficiencies, or movement deficiencies specific to each individual.

  • The system can analyze multiple athletes rapidly (US Soccer completed 120 analyses in 14 minutes) and is used weekly by major professional sports teams across most major team sports globally.

  • Movement analysis in natural environments is critical because injuries often occur during acceleration, deceleration, lateral movement, and direction changes—conditions that cannot be accurately measured in laboratory or static settings.

  • AI is an umbrella term encompassing multiple technologies including computer vision (using imaging/video), machine learning (using learning models), and large language models, which ViewMotion combines for comprehensive movement analysis.

  • Identifying and correcting movement asymmetries is essential for injury prevention, as imbalances create compounding negative effects and improper movement patterns often result from inadequate initial training or incomplete rehabilitation after injury.

  • Visual feedback through augmented reality overlays on video is crucial for athlete buy-in and learning, as numerical data alone from GPS and force plates is less effective for understanding what improvements are needed and why.

Starter Healthy Heart Membership

Transcript English

Welcome, my name is Dr Warrick Bishop. I'm a cardiologist, an author and a keynote speaker. I'm CEO of the Healthy Heart Network. I'm all about trying to help people live as well as possible for as long as possible. Heart disease is huge in Australia. Every 20 minutes someone suffers a heart attack. Most of these could probably have been avoided if only we knew what to do. This podcast is all about helping you understand. blood pressure, weight, cholesterol, for better health. If you enjoy this podcast, I would be honored for a five-star review. You can share it with your family and friends. It may well save someone you love. Hi, it's Warwick here, and thank you for joining me on my podcast and videocast station. As always, I'm really grateful if you're taking the time to listen to what I'm sharing. It means you're not listening to something else, and you've found this valuable in the past. I often say I'm super excited to speak with people, and everyone says they're super excited to speak with people, but I'm genuinely really pleased to catch up with today's guest, who's Ryan Talbot. He's president and founder of his own company called ViewMotion, and we're going to talk about it in a minute. But before we do, I'm going to share with you a short story as to how we got here, because it still makes me smile. I'll introduce Ryan first, but then I'll share. So hi, Ryan. Say g'day. Hi, thanks for having me on. So before we dive in, Ryan and I are here because my wife and I were traveling back from South America. We had a 12-hour layover in Santiago. We're sitting in an airport pub having a beer and some nice greasy food. And a chap comes by to plug in his computer and various bits and pieces. And through conversation, he was having a shell. I just turned around because I was sort of sitting at a nearby table and said, well, that sounds like an Australian accent. Anyway, before you knew it. Daniel, Ryan's brother and I were chatting away. We quickly got to, I'm a doctor, he's in AI, his brother's in AI. And before we knew it, Daniel had called Ryan on the phone. We were talking to each other and we were talking about the possibility of how AI movement interpretation could be brought to medical health and wellbeing. Fabulously interesting space. And I really am genuinely pleased and excited to be talking with Ryan about it today. So there's the intro, right? How do you feel about that? That's fantastic. Let me invite you to start off by just a very quick potted summary. I know you're from Sydney. I know you're in Houston, Texas these days. How did that journey happen? And tell us a tiny bit about ViewMotion, if you would. Yeah. So firstly, thanks for having me on. I'm certainly excited to talk about some of the things that we'll be talking about today. And yeah, quite uncanny how we did meet because my brother just said, hey, speak to Ryan, put me on the phone and we were chatting away. So here we are. So a little bit about sort of myself and I guess my journey with ViewMotion. So I spent... 25 years in the technology and security space and really looking at a lot of high-end technology with biometrics and identity. In fact, I actually brought the first Iris biometrics into Australia for the Commonwealth Bank back in 2003. Wow. I've been involved in that space for many years. I was actually working as a consultant at the Galaxy Casino in Macau, which is probably the largest in the world. And I invited all the camera manufacturers and their engineers into the casino to be part of what we call a shootout. And I have to be careful saying that in Texas. But so essentially what we're looking at doing is as we're going over from analog to digital and IP video, we'll look. looking at how we could really enhance the casino environment to be able to use video, but also AI and analytics as well. So that was sort of an area that I guess, yeah, we were really looking at how we could just drive that market with that technology. And about three months later, I ended up working at Canon. They were one of those companies. that were part of that evaluation of video technology and analytics. So then I ended up heading up the Canon's network video solutions and analytics division. But I was also a track and field coach for 15 years. So I've always been very interested in movement and really looking at how can we... you know, how can we change this space? Because video is the most used tool by everyone within sport and coaching, but there was really no easy way to be able to measure and understand how people move. So that's sort of how ViewMotion started. And then we started getting involved in sport and now we work with a lot of professional sporting leagues around the world. So for someone who isn't in the coaching space and doesn't understand how coaches are using videos with their advocate, someone like me actually. How does that information feed into coaching practices, right? Yeah, so I think if we look at how athletes move in their natural environment, and I guess this has been one of the challenges, right? Because we could measure things in a lab. We could measure things in the gym vertically, statically, or in isolation. But when we get out in the natural environment, it's very difficult to be able to accurately measure how people move. So as an example, how they accelerate, how they run, how they decelerate, and this is where a lot of injuries start occurring as well, how they change direction. how they move laterally, how they jump. So what we have done at ViewMotion is really looked at how we can calibrate a scene in a natural environment to be able to really analyze, deeply analyze what's happening basically from head to toe and also on the ground, to be able to create models off the back of that to understand based on even normative data sets, looking at the age, gender, sport, cohort, height, anthropometric values. to be able to really understand how they move and how they need to improve and also reduce the likelihood of injuries as well, which is really important in sport. And so that evaluation that ViewMotion does, is that done on a video and then post-processed or is it done in real time? Yeah, so it's captured from any smartphone and that's uploaded. through to the ViewMotion cloud. And then what that does is then turns video into data, data back onto video as augmented reality. And then from there, it creates kinograms. So kinograms are key points or positions of movement in time. And that's not something. Gram meaning picture. So a movement. Yeah. Yeah. And it was actually a funny story about this going back to the 1880s. So Leyland Stanford, as in Stanford University, had a bet with someone that a horse's feet are off the ground momentarily, but you can't see it with your eyes. So he employed a photographer, Edward Muybridge, to be able to rig up a series of cameras to be able to get that horse moving to be able to see if its feet were off the ground. I was fascinated by that. And I would spend hours every week trying to cut out shapes and understand how my athletes moved. From there, what we did was look at how can we use computer vision and machine learning to be able to do that around how humans move and to be able to understand those movements. So if I were a coach and I took a video of my athlete, I would upload it to your cloud and there would be a processing parameters that are undertaken. And that would generate movement graphs, kinographs. And that would give me and inform me as to some of the risks of injury or improvement opportunities in terms of performance. What are some of the things that those pieces of information would give you, right? Yeah, so it actually spits out a really nice report. So it gives you times, split times. It gives you maximum speed. It shows where you are against your all-time best. It gives you asymmetries between the key metrics of step length and step frequency, which makes up speed. And then... flight time, ground contact time. And then we can start looking at how we're moving at speed on one leg at a time, on left leg and right leg, and how do we move throughout time and space. And then from there, we look at a step quality score of looking at the kilograms and then looking at some of the key spatiotemporal parameters of how someone moves based on a normative data set. And then it gives you a progress over time. And it also spits out three things that you need to work on. So based on... algorithms that it says okay if you're deficient in this area you might be force deficient or velocity deficient or movement deficient it will provide three key things you need to work on and the recommendations off that and then also how to read your report as well and the video actually supports that by being able to visually show you how you're moving how you're connecting with the ground what shapes you make instantaneous speed times and those things so you can visually see that and from a sports perspective it's very important to get buy-in from the athlete as well because so many things give you numbers these days like gps and force plates and all these type of things but from a learning perspective it's really really important to actually understand what you're working towards and why yeah that makes sense so presumably the videos that you're obtaining are a standard format so you ask the coach to obtain 30 seconds of someone doing a particular thing is that presume is it a particular protocol that people need to follow as they take those videos for that processing yeah it's quite simple so we we calibrate the scene using five cones so you just mark that out and essentially what you're now doing is creating a lab out you know grass on the track on a court we also work on ice as well and From there, we put a camera on a tripod and we use 4K and 60 frames. So every phone has that now. yeah and that's important so 4k gives us 3840 horizontal pixels so a lot of pixels to be able to work with so essentially two pixels per centimeter of what's happening on the ground and what's happening with the body as well and it's mapping out all the anatomical points of the body and then from there based on the assessment i think we have about 26 different assessments now and from there basically the camera will triangulate those cones and then it will measure everything out to pretty well centimeter accuracy. And then once it's captured, so a good example of US soccer did a research paper, and I think it was about 120 analysis they did in about 14 minutes. So they can run a lot of people through it very quickly and teams use it every single week across every major sport, pretty much every major team sport. So that allows them to be able to get all of the information they need to be able to look at. how their athletes are moving and then if we look at from an injury perspective but there's a lot of research in this space around how someone's connecting with the ground and where their swing leg is and where forces are going into the body and if we can identify movement deficiencies then we can look at putting interventions in place to be able to improve that because a lot of people were just not taught how to move properly yeah and when we have a have an injury um sometimes we're not rehabilitated properly as well so that creates an imbalance and when we have an imbalance there's a compounding effect so now we're starting to look at if someone has a significant asymmetry we need to get that closer together so being able to understand again how they're moving at speed horizontally on one leg at a time we can't measure that vertically we can only measure it by using video we also have multiple camera systems so if you're doing counter movement jump it could be from the side and front together and we can start looking at things like knee valgus we can start looking at hip and trunk angles we can bring everything together to actually understand that movement wow okay so An absolute potpourri and a wealth of data that a coach can then come back and fine tune the athlete, not just for performance, but also for injury prevention, which is really powerful. But I think that's also a really nice segue because one of the things that... I mean, really what we started talking about when Daniel passed his phone to me with you on the other end was how can we use this amazing technology, which there's no question this is fantastic. And just let me check, is it AI driven or is it machine learning driven? And for those who are listening and not sure, my understanding is machine learning. is you have thousands and thousands of data sets and give that to a database and then the machine accesses all that information versus AI where there's a little bit more self-interpretation by the program as to what's going on. Is that a reasonable summation of that, Brian, or not? i think the best way to describe it is ai being artificial intelligence is an umbrella and that simply means that a human's not doing it that a machine's doing it then we look at computer vision so there are different areas of ai and if we look at computer vision that means we're using imaging or video so that's what the computer vision aspect is then we have machine learning and that's where we have a learning models to be able to understand what it is and what it means and those type of things so we'll use computer vision and machine learning together because we're using imaging and then we have things like large language models and that's another aspect of ai so there are many different areas of ai but artificial intelligence is an umbrella term so to speak if that makes sense Just for those listening, if you're curious, one of the things that might be a good example to understand machine learning is if you took hundreds and thousands of chest x-rays of normal people, fed them into a machine so that you created that resource, then you can show the machine that... uh another x-ray and it can compare it to you if it's normal or abnormal that would be a fair sort of example absolutely yeah yeah yeah for sure so so we're using this fantastic ai technology to uh interpret for athletes but the the thing that we started talking about was how can this be pivoted towards um diagnosis early diagnosis for individuals and i think i said to you that my interest from an observational perspective is that we run a wellness center called OsteoStrong. And part of our process is helping people with balance. We put them on vibration plates and we do different balance maneuvers. And as I watch these individuals, I know that there is early pathology there, but I don't have a mechanism to know how to deal with it. And so this sort of technology may bring... amazing opportunity to be evaluating even subtle changes at an early stage that could really change and inform what we do for individuals in their best health journey. Would you like to speak to that a bit? Yeah, definitely. Well, I think movement in general is important to all of us, to everyone through every stage of life. And although it's good for athletes when we're looking at improving performance and reducing injuries and return to training, return to performance, I think... As we start to get older, then how we move is really important because we start to slow down. And I spent many years in the surveillance space and basically on video, watching people, crowds of people. I didn't have any sound. So I studied movement and understanding how people moved. And I found it really interesting because the human brain doesn't actually recognize this. we actually adapt to our environment so there's something known as rhythm lock and we walk together we jog together we run together and sprint together lower socioeconomic environments actually walk slower and they synchronize together within that scene I've got some really good videos to actually show some of that which we can do at some stage but so when we start looking at things like motor patterning and the neuromuscular aspect of how we move I think you know we can start to really understand how someone's changing over time. And this has never been done before. So your walking gait is really, really powerful understanding of your health as well. You know, over the age of 60, your walking gait can quite accurately determine your life expectancy. And there's been a number of studies, one done out of Harvard. Again, it's been difficult to really break down and understand that within a natural environment because it wasn't technology to do that. We could do it in a lab. Now we can actually do that in a natural environment. And we look at things like falls, you know, one in three people over the age of, I think it's either 60 or 65, that fall and break a hip will die within the first 12 months. Movement's a big part of that as well. So if we can actually understand how people move and we can put and we're seeing changes and we look at different pathologies that could be identified, different diseases and pathologies that we can see these changes in these movement patterns because we're able to. really get detailed understanding of how that person moves and we've had some trials internally looking at some of this with you know some people in this space that have been sort of measuring it and have actually picked up on some things and got a second opinion and you know things like Parkinson's yeah that's a really good example. In fact I was going to suggest Parkinson's I was going to say as a medical student we had a fabulous professor of neurology who actually mimicked some of the walks of a stroke, of Parkinson's disease, of peripheral neuropathy, of cerebellar ataxia. And of course he did them and they're very obvious because they're very specific. But I reckon... And they're all different, right? So a stroke walk, legs stiff, circumnavigates as it moves forward. Parkinson's is slow, small steps and so forth. So these are all very clear characteristics. The really interesting thing is there's got to be this possibility to pick these things up really early. And as you were talking about it, just for those listening or watching. Think about how many things go into your gait. So Ryan was saying it's such an important indicator of your general wellbeing. Well, of course it is. It's reflective of your muscular strength for a start. And we know muscular strength is hugely important for vitality, longevity, independence. So strength is important. Your nerves are all tied up in there. So you can have peripheral neuropathy. You can have a single nerve affected. You can have the brain affecting the nerves. Your balance, cerebellar is affected by nerves. Sensation, proprioception is affected by nerves. Your joints could be affected and impact your gait. And it goes on. This is really very complicated stuff. And if you could tease some of those individual components out, particularly in an early stage, it would be extraordinary. Absolutely. And I think, you know, if we look at sort of how we optimize, you know, how we move and I love the S's. So there's a strength component. There's a skill component of how we move. Stability is a big one. Symmetry of movement, suppleness or. you know looking at flexibility and stamina so looking at sort of all the s's there's a few more as well um you know psychology but uh yeah so being able to understand each aspect of of those s's and being able to optimize that and being able to measure it it's uh we're at a great point in time at the moment where we can measure things really well and understand who we are and what we need to improve and if we want to improve those things we can And, you know, we can get regular blood tests. We can get movement assessments now. There's so many things we can do because we're in charge of our health that it's important to be able to have the right tools to be able to do that. Ryan, tell me, you're obviously in the sport and athlete space at the moment, but are you starting to accumulate data or look towards diagnostics in the medical sphere and how might that look? What do you have to do to make that happen and what sort of timeline would be behind it if you were heading in that direction? Look, that's a really good question. And, you know, over the last few years, it was really about let's look at something that, you know, what are the best and fastest humans on earth do? And we now have a data set of over 100,000 mainly professional and semi-professional athletes over time. So that's a really, really good data set. And in fact, from a biomechanical standpoint across the most amount of sports, it's the largest in the world. And again, It's really looking at, and as I said earlier, movement's important to all of us. And we have a number of surgeons, orthopedic surgeons, a neurospine surgeon that's involved in view motion, saw what we're doing earlier on. The regulatory pathway is obviously very different to sport, and that's what we're working towards now. But, you know, we've got a, you know, that type of data set is absolute gold because we can really start to use. big data to understand what's going on. So it's a focus we're looking at at the moment. And, you know, we've got some great people behind it. It's sports great, but I think general population, aging population that we can make a huge difference towards. So over the next 12 to 18 months, we're heading down that space. And I want to get to a point where I'm analyzing 1 million people a day to be able to improve their lives. That's the ultimate goal. Wow, powerful, man. It's really powerful. Yeah, yeah. And I reckon as you think more and more about it, gate is such a, almost a canary in the mineshaft. Once that gate starts to be affected, you know you need to be looking beyond to find out what's happening. It's a very sensitive indicator, so extraordinary. Yeah, definitely. Look, I think we'll wrap up. Was there anything else you wanted to share, Ryan, before we go? This has been just fascinating. I love this sort of stuff, this technology, what you're doing at the moment, where it might go. I really hope that the people listening get as much out of it as I have. I can see you're deeply passionate about it and I'd love to see where it all goes in the next couple of years because it just does my mind in. I think it's fantastic. But anything else you wanted to share? Yeah, I was just going to say, look, we've got an incredible team working behind ViewMotion that works seven days a week. We're really changing a lot of things for a lot of people. And if anyone listening wants to get involved, either from a business perspective or even from an investment perspective, we're looking for people to come on board to be part of this. And it's an exciting journey. So yeah, anyone listening that has an interest in this space, in the future of human movement, absolutely. What about a research space as well? So business space, investment space, and a research space, you must be interested in some people who could contribute to that as well, I'm guessing. Absolutely. Look, we're collaborating with a number of universities at the moment. I think we have about 15 research things going on at the moment. We've got two in the English Premier League. We've got US Soccer have done one. We've got two out of Australian Catholic University, Melbourne University, also Auckland. Yeah, so all over the world. It's good to see what we're doing in sports space, but research space definitely within. Health is an area that we would certainly like to do a lot more in and just getting good people behind it. Yeah. No, brilliant. Ryan, all the way from you're in Austin, Texas, aren't you? Austin, yes. I know you were traveling around a bit and we had to shift our catch up, but currently in Austin, Texas, where he's based. Thank you so much for sharing. I really do appreciate your time and making the effort to catch up because. We did have to get our diaries aligned. I'd like to thank your brother for putting us in touch. Absolute weapon and a real people person who in minutes had you on the phone. For those listening, I hope you found today as interesting as I have. It really is fascinating space. As always, if you've got any queries or questions, drop us a line at info at drorichbishop.online. And again, appreciate you listening. Till next time, I hope you live as well as possible for as long as possible. Take care and bye for now. Did you know that coronary artery disease kills one in four people? So most of us are likely to carry some risk or know someone who does. If you're interested in finding out more about how to evaluate that risk, check out www.virtualheartcheck.com.au It'll give you information about risk and what else can be done to be even more precise.