1 00:00:08,240 --> 00:00:19,440 Speaker 1: Hushkin, what's one surprising thing your work has taught you 2 00:00:19,600 --> 00:00:21,079 Speaker 1: about elite athletes? 3 00:00:22,160 --> 00:00:24,680 Speaker 2: I never thought I would see muscles that were so 4 00:00:25,560 --> 00:00:30,520 Speaker 2: developed they broke our scale. Wow. 5 00:00:31,960 --> 00:00:34,159 Speaker 1: Yeah, like it was just too big the machine, the 6 00:00:34,240 --> 00:00:35,839 Speaker 1: AI couldn't figure out what it is. 7 00:00:35,960 --> 00:00:38,159 Speaker 2: Well, no, the AI found it, but we are like 8 00:00:38,240 --> 00:00:39,600 Speaker 2: our kind of rating system. 9 00:00:40,200 --> 00:00:43,280 Speaker 1: Wow. Was there a particular athlete or a particular sport 10 00:00:43,360 --> 00:00:44,400 Speaker 1: or particular muscle? 11 00:00:44,520 --> 00:00:44,680 Speaker 2: What? 12 00:00:44,680 --> 00:00:46,320 Speaker 1: What? What muscle broke the scale? 13 00:00:47,400 --> 00:00:49,720 Speaker 2: Uh? The gluteus maximus breaks it A. 14 00:00:49,760 --> 00:00:53,040 Speaker 1: Fair kidding fantastic. 15 00:00:53,120 --> 00:00:54,639 Speaker 2: Yes, it's a pain in my butt. 16 00:00:56,800 --> 00:00:58,360 Speaker 1: Like because it's too big. 17 00:00:58,880 --> 00:01:00,720 Speaker 2: Yeah, it's just so big. 18 00:01:06,680 --> 00:01:09,440 Speaker 1: I'm Jacob Goldstein and this is what's your Problem? This 19 00:01:09,600 --> 00:01:12,520 Speaker 1: month a bunch of Pushkin podcasts are coming out with 20 00:01:12,680 --> 00:01:17,839 Speaker 1: Olympics inspired shows. Revisionist History has a series about America's 21 00:01:17,840 --> 00:01:22,640 Speaker 1: decision to participate in Hitler's Berlin Olympics in nineteen thirty six. 22 00:01:23,200 --> 00:01:25,960 Speaker 1: The Happiness Lab has an interview with a coach who 23 00:01:26,000 --> 00:01:29,959 Speaker 1: coaches coaches and here on What's Your Problem, We're going 24 00:01:30,040 --> 00:01:32,200 Speaker 1: to be talking with people who are working at the 25 00:01:32,240 --> 00:01:38,560 Speaker 1: frontiers of technology to help elite athletes perform better. For example, 26 00:01:39,360 --> 00:01:43,760 Speaker 1: today my guest is Sylvia Blemker. She's a professor of 27 00:01:43,800 --> 00:01:47,760 Speaker 1: biomechanical engineering at the University of Virginia, and she's the 28 00:01:47,800 --> 00:01:52,200 Speaker 1: co founder of a company called Springbok Analytics. Sylvia's problem 29 00:01:52,320 --> 00:01:55,880 Speaker 1: is this, how do you combine MRI scans and artificial 30 00:01:55,920 --> 00:01:59,560 Speaker 1: intelligence to generate new insights that can help both elite 31 00:01:59,600 --> 00:02:03,440 Speaker 1: athletes and people suffering from diseases that affect the muscles. 32 00:02:04,280 --> 00:02:09,400 Speaker 1: Springbox clients include medical researchers, Olympic athletes, major League baseball, 33 00:02:09,760 --> 00:02:13,920 Speaker 1: and several professional basketball and soccer teams. You'll hear about 34 00:02:13,960 --> 00:02:16,440 Speaker 1: all that on the show, but first we're going to 35 00:02:16,520 --> 00:02:20,000 Speaker 1: pick up where we left off in the conversation. We 36 00:02:20,000 --> 00:02:25,480 Speaker 1: were discussing the extraordinarily large muscles of elite athletes, and 37 00:02:25,520 --> 00:02:28,840 Speaker 1: then Sylvia told me something even more surprising. 38 00:02:30,000 --> 00:02:32,960 Speaker 2: The other thing is that they have some tiny muscles too, Huh. 39 00:02:33,360 --> 00:02:36,839 Speaker 1: Like they have like smaller than a normal person's muscle. 40 00:02:36,880 --> 00:02:37,440 Speaker 2: Much smaller. 41 00:02:37,560 --> 00:02:38,040 Speaker 1: Huh. 42 00:02:38,120 --> 00:02:39,639 Speaker 2: They put their muscle where they need it. 43 00:02:40,120 --> 00:02:42,640 Speaker 1: What's an example, Like what muscle is tiny and what 44 00:02:42,720 --> 00:02:43,440 Speaker 1: kind of athletes? 45 00:02:44,320 --> 00:02:49,239 Speaker 2: Calf muscles are small in most fast athletes huh? And 46 00:02:49,360 --> 00:02:53,399 Speaker 2: ether you look at a sprinter or like a running back. 47 00:02:53,960 --> 00:02:57,279 Speaker 1: It's just all quad no calf, all. 48 00:02:57,040 --> 00:03:00,880 Speaker 2: Like thigh no calf, yeah, THI And it kind of 49 00:03:00,880 --> 00:03:03,000 Speaker 2: makes sense because you know, if you're trying to run fast, 50 00:03:03,000 --> 00:03:04,840 Speaker 2: you wouldn't want to put a lot of mass like 51 00:03:04,960 --> 00:03:07,240 Speaker 2: at the end of your leg. It's like as a 52 00:03:07,240 --> 00:03:08,959 Speaker 2: lot of inertia to like move your leg. 53 00:03:09,240 --> 00:03:09,520 Speaker 1: Huh. 54 00:03:09,960 --> 00:03:12,880 Speaker 2: Because you know, the muscles are important for sprinting, that's 55 00:03:12,919 --> 00:03:16,440 Speaker 2: the interesting thing, but they just don't they're small, very. 56 00:03:16,600 --> 00:03:20,520 Speaker 1: H huh uh huh. So I'm particularly interested at this 57 00:03:20,680 --> 00:03:27,679 Speaker 1: moment in the sports piece of what you do. I'm curious, 58 00:03:27,800 --> 00:03:31,160 Speaker 1: by the way. Do you work with any Olympic teams 59 00:03:31,240 --> 00:03:33,520 Speaker 1: or Olympic athletes? Yeah? 60 00:03:33,639 --> 00:03:38,040 Speaker 2: Yeah, We've actually been working with several different Olympic athletes. 61 00:03:39,440 --> 00:03:42,000 Speaker 2: The ones that probably that come to mind most are 62 00:03:42,160 --> 00:03:45,760 Speaker 2: multiple players on the US women's national soccer team. 63 00:03:46,000 --> 00:03:48,720 Speaker 1: Oh cool, tell me, like, tell me the story of 64 00:03:48,760 --> 00:03:52,440 Speaker 1: that of that work. So they came to you, what 65 00:03:52,480 --> 00:03:54,160 Speaker 1: did what did they what do they want when they 66 00:03:54,160 --> 00:03:55,800 Speaker 1: came to you? Like, how did that? How did that begin? 67 00:03:57,840 --> 00:04:01,240 Speaker 2: They came to us along with their team. So the 68 00:04:01,280 --> 00:04:04,200 Speaker 2: technology we provide, you know, an athlete could understand it, 69 00:04:04,240 --> 00:04:07,760 Speaker 2: but really with their team to help them figure out 70 00:04:07,840 --> 00:04:10,120 Speaker 2: how to keep athletes healthy. 71 00:04:10,320 --> 00:04:12,920 Speaker 1: So what did they what did they say? What did 72 00:04:12,960 --> 00:04:13,960 Speaker 1: they say when they came to. 73 00:04:14,880 --> 00:04:19,679 Speaker 2: So, for example, one athlete that's coming to mind had 74 00:04:20,560 --> 00:04:24,960 Speaker 2: a known imbalance side to side that based on a 75 00:04:25,120 --> 00:04:28,400 Speaker 2: history of injury, and they really wanted to know where 76 00:04:28,400 --> 00:04:29,960 Speaker 2: that imbalance was coming from. 77 00:04:30,080 --> 00:04:32,960 Speaker 1: So the woman had had hurt one of her legs, 78 00:04:33,000 --> 00:04:35,480 Speaker 1: and that leg was even after she came back, that 79 00:04:35,560 --> 00:04:38,320 Speaker 1: leg was weaker essentially than the other. I mean, is 80 00:04:38,360 --> 00:04:40,960 Speaker 1: that the sort of gross you know, macro. 81 00:04:41,040 --> 00:04:43,520 Speaker 2: Way, Yeah, exactly, that's a that's a nice way to 82 00:04:43,520 --> 00:04:43,720 Speaker 2: put it. 83 00:04:43,800 --> 00:04:47,240 Speaker 1: Yeah, And and they wanted a sort of finder like, okay, 84 00:04:47,240 --> 00:04:49,359 Speaker 1: but we can see that, but what's going on on 85 00:04:49,400 --> 00:04:53,200 Speaker 1: the inside, like muscle by muscle tell us that, yes. 86 00:04:53,080 --> 00:04:55,800 Speaker 2: Exactly, that's precisely what we do. We go on the 87 00:04:55,839 --> 00:05:00,760 Speaker 2: inside because on the outside you see perhaps that her 88 00:05:01,120 --> 00:05:04,640 Speaker 2: knee extents or quads seem weaker on one side than 89 00:05:04,680 --> 00:05:08,719 Speaker 2: the other. But there's four quads, quadre steps, four muscles, 90 00:05:09,160 --> 00:05:11,400 Speaker 2: and so it's not clear which of those muscles are 91 00:05:11,440 --> 00:05:15,839 Speaker 2: actually the culprit for that imbalance and in what way. 92 00:05:15,560 --> 00:05:19,760 Speaker 1: Good So this is their question, and then what happens next? 93 00:05:20,160 --> 00:05:23,960 Speaker 2: So This first step is an MRI scan, and so 94 00:05:24,960 --> 00:05:32,000 Speaker 2: with these athletes or teams, we have ways to connect 95 00:05:32,040 --> 00:05:35,480 Speaker 2: them with an MRI machine, whether it be through an 96 00:05:35,480 --> 00:05:39,400 Speaker 2: imaging center that they partner with, or we've even actually 97 00:05:39,400 --> 00:05:45,440 Speaker 2: brought MRI mobile trucks to sites to make it. 98 00:05:45,040 --> 00:05:47,760 Speaker 1: Like players run off the field and get an MRI 99 00:05:48,000 --> 00:05:51,200 Speaker 1: and go back and keep playing. Yeah, yeah, kind yeah, yeah. 100 00:05:51,520 --> 00:05:54,440 Speaker 2: It helps just with the timing of things. But so 101 00:05:54,520 --> 00:05:57,960 Speaker 2: first we connect them there, so it takes about ten minutes. 102 00:05:58,560 --> 00:06:03,000 Speaker 2: Then they send those pictures up into the cloud into 103 00:06:03,080 --> 00:06:06,560 Speaker 2: our server and then we crunched through it and then 104 00:06:06,640 --> 00:06:10,680 Speaker 2: we send back a report on their muscles. We also 105 00:06:10,920 --> 00:06:14,799 Speaker 2: have what we call it interactive Viewer, and it's presented 106 00:06:14,880 --> 00:06:18,520 Speaker 2: in the form of a three D model. Three dimensional model, 107 00:06:18,560 --> 00:06:22,120 Speaker 2: so you actually see your own legs, the muscles and bones, 108 00:06:22,200 --> 00:06:26,280 Speaker 2: your own muscles and bones that we've identified from the 109 00:06:26,320 --> 00:06:30,960 Speaker 2: images going through a process called segmentation where we find 110 00:06:31,000 --> 00:06:33,160 Speaker 2: all the muscles and bones and then we reconstruct them, 111 00:06:33,160 --> 00:06:35,040 Speaker 2: so it's kind of like a digital twin of that 112 00:06:35,080 --> 00:06:38,479 Speaker 2: person that they can see on their computer. And so 113 00:06:38,680 --> 00:06:41,240 Speaker 2: along with it or a number or all these metrics 114 00:06:41,320 --> 00:06:47,880 Speaker 2: that helps them understand their balance. The development or strength 115 00:06:47,920 --> 00:06:50,120 Speaker 2: of the muscles and the health of the muscles. 116 00:06:50,520 --> 00:06:52,880 Speaker 1: So tell me about this report, they get like, what 117 00:06:52,920 --> 00:06:54,360 Speaker 1: does it? What does it say? 118 00:06:55,960 --> 00:06:57,920 Speaker 2: So the basis of that is actually a lot of 119 00:06:57,960 --> 00:07:01,479 Speaker 2: research that we did over many years, because you need 120 00:07:01,520 --> 00:07:06,400 Speaker 2: to understand where somebody falls relative to a normal essentially 121 00:07:06,440 --> 00:07:09,320 Speaker 2: to give them. Essentially, we have a scoring system for 122 00:07:09,560 --> 00:07:13,200 Speaker 2: the muscles and that's based on comparing with a large 123 00:07:13,280 --> 00:07:18,440 Speaker 2: data set of healthy individuals. And so we know for 124 00:07:18,480 --> 00:07:23,920 Speaker 2: a given person, based on their sex, age, height and weight, 125 00:07:24,680 --> 00:07:27,280 Speaker 2: how big we expect all the muscles to be. And 126 00:07:27,760 --> 00:07:30,320 Speaker 2: that's through a lot of previous research. So then we 127 00:07:30,360 --> 00:07:34,120 Speaker 2: can say, okay, here's where you land each particular muscle 128 00:07:34,520 --> 00:07:38,240 Speaker 2: compared to this nor what we call a normative database, 129 00:07:38,840 --> 00:07:40,520 Speaker 2: so we call it a spring box score. 130 00:07:40,880 --> 00:07:42,560 Speaker 1: Do you do it for every muscle in the leg. 131 00:07:42,480 --> 00:07:46,360 Speaker 2: Or we do it? So our primary product that we 132 00:07:46,400 --> 00:07:50,240 Speaker 2: started with was every muscle in the legs essentially from 133 00:07:50,320 --> 00:07:54,520 Speaker 2: belly to feet, no muscle left behind. They're all important. 134 00:07:54,720 --> 00:07:56,400 Speaker 1: How many muscles are there. 135 00:07:56,960 --> 00:07:58,120 Speaker 2: Thirty five per leg? 136 00:07:58,200 --> 00:08:01,800 Speaker 1: So seven okay, okay, yeah, more than I would have guessed, 137 00:08:01,960 --> 00:08:07,840 Speaker 1: but a lot. Yeah, all important, okay. So, and so 138 00:08:08,080 --> 00:08:12,000 Speaker 1: it's basically, how strong and healthy is every one of 139 00:08:12,040 --> 00:08:15,800 Speaker 1: those seventy muscles relative to baseline? 140 00:08:16,080 --> 00:08:19,960 Speaker 2: Right, And then the asymmetry comes where you can compare 141 00:08:20,080 --> 00:08:22,280 Speaker 2: side to side. So for each of the thirty five 142 00:08:22,360 --> 00:08:24,920 Speaker 2: muscles that they exist on each leg, we can say 143 00:08:25,520 --> 00:08:27,960 Speaker 2: which side is bigger, which side is smaller, and by 144 00:08:28,000 --> 00:08:31,280 Speaker 2: how much? And then we also have normative values for 145 00:08:31,400 --> 00:08:34,080 Speaker 2: that because we're all just slightly asymmetric. 146 00:08:34,320 --> 00:08:38,120 Speaker 1: Uh huh, And presumably some muscles are more asymmetric than others, 147 00:08:38,160 --> 00:08:41,000 Speaker 1: And so you want to know kind of how how 148 00:08:41,120 --> 00:08:46,880 Speaker 1: asymmetric relative to baseline is this particular pair of muscles exactly? Yeah, 149 00:08:47,360 --> 00:08:50,520 Speaker 1: And and so in the in the case of this 150 00:08:50,880 --> 00:08:53,920 Speaker 1: uh soccer player who came to you who you know, 151 00:08:54,120 --> 00:08:56,560 Speaker 1: knew knew she had some kind of problem with her 152 00:08:56,640 --> 00:08:58,839 Speaker 1: quadriceps on one side, but didn't know what was going on. 153 00:08:59,320 --> 00:09:00,200 Speaker 1: What did you find? 154 00:09:01,640 --> 00:09:05,880 Speaker 2: We found some imbalances, and actually not just in those muscles. 155 00:09:06,640 --> 00:09:10,080 Speaker 2: It turns out that, you know, it's all connected. So 156 00:09:11,679 --> 00:09:14,679 Speaker 2: if you have a weakness or an imbalance and one 157 00:09:14,720 --> 00:09:17,000 Speaker 2: set of muscles, usually some other set of muscles are 158 00:09:17,000 --> 00:09:18,200 Speaker 2: compensating in someone. 159 00:09:18,360 --> 00:09:21,240 Speaker 1: Yeah, well, it's like when you like mess up. Even 160 00:09:21,280 --> 00:09:24,600 Speaker 1: if you're just a recreational athlete, right Like, if you 161 00:09:24,640 --> 00:09:26,800 Speaker 1: mess up something, you mess up your ankle, then you 162 00:09:26,840 --> 00:09:29,280 Speaker 1: start walking funny, and then like your back hurts because 163 00:09:29,280 --> 00:09:32,240 Speaker 1: you're walking funny. Right Like, that is a very anecdotally 164 00:09:32,240 --> 00:09:33,000 Speaker 1: apparent thing. 165 00:09:33,080 --> 00:09:35,160 Speaker 2: Yeah, yeah, we all know that, but that you know 166 00:09:35,240 --> 00:09:37,440 Speaker 2: it shows through in the skin. But the thing is 167 00:09:37,480 --> 00:09:41,600 Speaker 2: that it's not very intuitive from the outside which muscles 168 00:09:41,720 --> 00:09:45,520 Speaker 2: have been affected, how they've compensated, and it looks different 169 00:09:45,559 --> 00:09:46,680 Speaker 2: for every single person. 170 00:09:46,920 --> 00:09:47,439 Speaker 1: Huh. 171 00:09:47,480 --> 00:09:50,280 Speaker 2: So that's why the report is very valuable because for 172 00:09:50,400 --> 00:09:54,360 Speaker 2: that person, they know exactly which muscles are the ones 173 00:09:54,400 --> 00:09:56,360 Speaker 2: that they really need to target, both the ones that 174 00:09:56,400 --> 00:09:59,360 Speaker 2: they already thought maybe were an issue, but then all 175 00:09:59,360 --> 00:10:03,160 Speaker 2: the other ones that showed up and they didn't really realize. 176 00:10:03,240 --> 00:10:05,040 Speaker 1: And so in the case of this soccer player, was 177 00:10:05,080 --> 00:10:09,720 Speaker 1: it like one particular quadrcep on one side that was 178 00:10:09,800 --> 00:10:11,560 Speaker 1: like the core thing and you could figure out which 179 00:10:11,559 --> 00:10:12,040 Speaker 1: one it was. 180 00:10:12,440 --> 00:10:14,200 Speaker 2: There was a few mess It wasn't just that. I 181 00:10:14,200 --> 00:10:18,480 Speaker 2: think there were at least one calf muscle and then 182 00:10:18,679 --> 00:10:22,319 Speaker 2: some in especially in the deep hip, those were impacted. 183 00:10:23,240 --> 00:10:26,199 Speaker 2: So yeah, it kind of shows up everywhere. 184 00:10:26,679 --> 00:10:30,000 Speaker 1: And so you have this essentially diagnosis, right, a very 185 00:10:30,040 --> 00:10:33,679 Speaker 1: sort of fine grained kind of diagnosis. Do you also 186 00:10:33,840 --> 00:10:35,920 Speaker 1: have a have a prescription? Do you have sort of 187 00:10:36,240 --> 00:10:41,280 Speaker 1: particular kinds of training to address these very fine grained things, 188 00:10:41,360 --> 00:10:44,040 Speaker 1: or do you leave that to the trainers or whoever? 189 00:10:44,600 --> 00:10:47,080 Speaker 2: We leave that to the trainers, because I think that 190 00:10:47,160 --> 00:10:49,800 Speaker 2: it's also important to have all the other information about 191 00:10:49,840 --> 00:10:53,319 Speaker 2: the athlete. We're not arguing that it replaces everything else. 192 00:10:53,640 --> 00:10:57,280 Speaker 2: And people pair it with lots of different other types 193 00:10:57,320 --> 00:11:02,320 Speaker 2: of measurements depending on the application or in the setting, 194 00:11:02,400 --> 00:11:06,000 Speaker 2: Like some people pair it with let's say, metrics of 195 00:11:06,720 --> 00:11:11,120 Speaker 2: jump performance. I'm shifting over to basketball here, but that's 196 00:11:11,160 --> 00:11:12,959 Speaker 2: just one that came to mind, where you can look 197 00:11:12,960 --> 00:11:16,680 Speaker 2: at the asymmetry about how how an athlete jumps, but 198 00:11:16,720 --> 00:11:18,760 Speaker 2: then you can also compare it to the asymmetry of 199 00:11:18,760 --> 00:11:22,840 Speaker 2: their muscles and get some insight. So it definitely, you know, 200 00:11:23,080 --> 00:11:24,640 Speaker 2: plugs in with a lot of other things. 201 00:11:25,000 --> 00:11:29,760 Speaker 1: And to what extent can trainers or you know, strength 202 00:11:29,840 --> 00:11:36,680 Speaker 1: coaches develop programs that are sufficiently kind of fine grained 203 00:11:36,679 --> 00:11:39,840 Speaker 1: to match the kind of fine grained findings you're having, right, Like, 204 00:11:39,880 --> 00:11:42,920 Speaker 1: for example, if you find, as I understand you did 205 00:11:43,400 --> 00:11:48,120 Speaker 1: that a soccer player has one particular quadricep that is weak. Like, 206 00:11:48,400 --> 00:11:51,120 Speaker 1: are there workouts that target a single quadricep and not 207 00:11:51,160 --> 00:11:51,600 Speaker 1: the others? 208 00:11:52,120 --> 00:11:52,800 Speaker 2: Yep, there are. 209 00:11:53,200 --> 00:11:57,240 Speaker 1: That's cool for whichever quadricepp you're just like, just for fun, 210 00:11:57,320 --> 00:11:58,160 Speaker 1: give me an example. 211 00:11:59,520 --> 00:12:01,840 Speaker 2: You know. One one way that it's very simple is 212 00:12:01,920 --> 00:12:07,480 Speaker 2: using something called biofeedback. Huh. So you can measure whether 213 00:12:07,600 --> 00:12:11,360 Speaker 2: you use something called EMG, which is a way to 214 00:12:11,360 --> 00:12:14,199 Speaker 2: measure how much electrical activity is a muscle, and then 215 00:12:14,240 --> 00:12:17,280 Speaker 2: you can see which muscles you're using for a given task. 216 00:12:17,840 --> 00:12:20,120 Speaker 2: So if you give people the feedback of which of 217 00:12:20,120 --> 00:12:23,640 Speaker 2: those muscles they're using and say, oh, no, you're not 218 00:12:23,760 --> 00:12:26,880 Speaker 2: using this one, use this one more, that actually works 219 00:12:27,000 --> 00:12:27,760 Speaker 2: very effectively. 220 00:12:27,920 --> 00:12:32,280 Speaker 1: Oh really, So you can basically use your brain if 221 00:12:32,320 --> 00:12:36,760 Speaker 1: you're getting the feedback to focus on which quadricep you're training. 222 00:12:37,000 --> 00:12:39,880 Speaker 2: Yeah, and there's other ways you can give the feedback 223 00:12:39,880 --> 00:12:42,240 Speaker 2: in other different ways, but yeah, our brains are very 224 00:12:42,600 --> 00:12:44,880 Speaker 2: good at that. Once they get feedback, they're very good 225 00:12:44,920 --> 00:12:45,520 Speaker 2: at learning. 226 00:12:46,160 --> 00:12:49,080 Speaker 1: That's cool, especially somehow to think of what elead athletes right, 227 00:12:49,080 --> 00:12:52,080 Speaker 1: because they are already presumably like super dialed in in 228 00:12:52,160 --> 00:12:54,920 Speaker 1: terms of like the relationship between their brain and their 229 00:12:54,960 --> 00:12:57,120 Speaker 1: body at this very elite level exactly. 230 00:12:57,520 --> 00:13:00,920 Speaker 2: Yeah. The other I was going to mention a lot 231 00:13:00,960 --> 00:13:05,199 Speaker 2: of players and teams use this not just one time, 232 00:13:05,280 --> 00:13:09,640 Speaker 2: but overtime. So they'll get a scan, figure out a plan, 233 00:13:10,760 --> 00:13:14,400 Speaker 2: work on that for maybe three months or six months, 234 00:13:14,440 --> 00:13:16,559 Speaker 2: and then do another scan and see how things are 235 00:13:16,559 --> 00:13:20,839 Speaker 2: progressing and adjust accordingly. So that's definitely another way to 236 00:13:21,000 --> 00:13:24,160 Speaker 2: in the long term see if what they're doing is 237 00:13:24,240 --> 00:13:26,600 Speaker 2: resulting in the change that they're hoping to see. 238 00:13:27,280 --> 00:13:29,839 Speaker 1: So, what happened with that soccer player who had the 239 00:13:30,120 --> 00:13:33,960 Speaker 1: weak quadricep and other related Yeah, No, I. 240 00:13:34,000 --> 00:13:39,800 Speaker 2: Think she's doing great, like staying healthy and and you know, 241 00:13:39,880 --> 00:13:40,480 Speaker 2: getting ready. 242 00:13:41,000 --> 00:13:42,600 Speaker 1: Yeah. So I know you can't tell us her name, 243 00:13:42,640 --> 00:13:47,120 Speaker 1: but will we see her in the Olympics this time? Great? So, 244 00:13:47,240 --> 00:13:48,800 Speaker 1: as you were talking about that, I mean there was 245 00:13:48,840 --> 00:13:51,319 Speaker 1: a moment where it was like, Okay, the athlete goes 246 00:13:51,320 --> 00:13:55,120 Speaker 1: and gets the MRI and then you get the scan. 247 00:13:55,320 --> 00:13:57,400 Speaker 1: You get the scan, and then you said, like, you 248 00:13:57,440 --> 00:13:59,440 Speaker 1: crunch through the numbers and then you make the report. 249 00:14:00,000 --> 00:14:03,960 Speaker 1: Ssumably you crunching through the numbers is like the result 250 00:14:04,000 --> 00:14:06,120 Speaker 1: of many many years of work and kind of the 251 00:14:06,120 --> 00:14:08,760 Speaker 1: core of what your company does. So I want to 252 00:14:08,800 --> 00:14:10,920 Speaker 1: talk a little bit more about that and kind of 253 00:14:10,960 --> 00:14:15,000 Speaker 1: how you how you got here, Like how did you 254 00:14:15,040 --> 00:14:16,040 Speaker 1: come to start the company? 255 00:14:16,440 --> 00:14:21,840 Speaker 2: Mm hmm? How long do I have a while? 256 00:14:22,000 --> 00:14:23,680 Speaker 1: I mean it's not you know, it's not the radio, 257 00:14:23,720 --> 00:14:24,520 Speaker 1: it's a podcast. 258 00:14:25,360 --> 00:14:27,720 Speaker 2: You. I am a professor too, so I can go on. 259 00:14:27,920 --> 00:14:29,960 Speaker 1: Let me ask you this, What was the moment when 260 00:14:29,960 --> 00:14:31,600 Speaker 1: you decided to start the company? 261 00:14:32,880 --> 00:14:35,520 Speaker 2: I can give you one moment and then we can 262 00:14:36,080 --> 00:14:36,720 Speaker 2: do a. 263 00:14:36,640 --> 00:14:42,640 Speaker 1: Couple of moments. Yeah, that's what's what a good story is, 264 00:14:42,680 --> 00:14:44,080 Speaker 1: like three moments. 265 00:14:43,720 --> 00:14:45,680 Speaker 2: In time, right right? Three not four? 266 00:14:46,440 --> 00:14:48,360 Speaker 1: Four is a tricky number. We could do five, or 267 00:14:48,360 --> 00:14:52,160 Speaker 1: we could do one. I think I think is better. 268 00:14:52,600 --> 00:14:55,320 Speaker 2: Yeah. Yeah. So I'm a professor. I run a lab 269 00:14:56,000 --> 00:14:59,960 Speaker 2: and form my entire career. I've been fascinated with must 270 00:15:00,960 --> 00:15:05,680 Speaker 2: and how it works, and fascinated by something that we 271 00:15:05,800 --> 00:15:09,640 Speaker 2: all in like the muscle field. We call form function relationships. 272 00:15:10,280 --> 00:15:13,080 Speaker 2: So the idea that the way a muscle is shaped 273 00:15:13,600 --> 00:15:15,760 Speaker 2: and the way it's structured and how big it is 274 00:15:16,360 --> 00:15:20,000 Speaker 2: influences how well it works or how well it functions, 275 00:15:20,040 --> 00:15:22,760 Speaker 2: how strong it is, how well it behaves. And there's 276 00:15:22,800 --> 00:15:24,640 Speaker 2: a lot too there, and there's a lot of nuance. 277 00:15:24,680 --> 00:15:27,480 Speaker 2: And that's like I've spent a career studying that in 278 00:15:27,520 --> 00:15:30,520 Speaker 2: lots of different ways. So I've always been interested in 279 00:15:30,600 --> 00:15:33,520 Speaker 2: quantifying muscle and figuring out how that influence is, how 280 00:15:33,520 --> 00:15:36,800 Speaker 2: it works, and both in healthy people or in athletes, 281 00:15:36,920 --> 00:15:41,800 Speaker 2: and also in different patient populations different In particular, I 282 00:15:41,880 --> 00:15:47,400 Speaker 2: have an interest in movement disorders, so neuromuscular diseases that 283 00:15:47,560 --> 00:15:53,400 Speaker 2: lead to impairments and mobility and movement ability. So one 284 00:15:53,400 --> 00:15:56,400 Speaker 2: of the light bulb moments for this was the fact 285 00:15:56,440 --> 00:15:59,600 Speaker 2: that I'd been using MRI to study muscle in my 286 00:15:59,640 --> 00:16:03,680 Speaker 2: research for a long time. It's kind of a ubiquitous 287 00:16:03,720 --> 00:16:09,080 Speaker 2: tool or like often use tool and research. But I 288 00:16:09,160 --> 00:16:14,160 Speaker 2: was struck by the fact that I was hearing from, 289 00:16:14,200 --> 00:16:18,320 Speaker 2: in particular a surgeon collaborator, and the surgeon was telling 290 00:16:18,360 --> 00:16:21,920 Speaker 2: me about his work and helping kids with cerebral palsy 291 00:16:22,080 --> 00:16:26,280 Speaker 2: improve their movement where they had hindered movement and largely 292 00:16:26,400 --> 00:16:31,320 Speaker 2: because their muscles are impacted. Not only do they have 293 00:16:32,160 --> 00:16:35,680 Speaker 2: an impaired ability to control their muscles, but their muscles 294 00:16:35,800 --> 00:16:38,840 Speaker 2: end up with impairments in their structure or their form, 295 00:16:39,440 --> 00:16:42,680 Speaker 2: which then influences how well they work. So surgeons have 296 00:16:42,760 --> 00:16:44,920 Speaker 2: to go in and do surgeries to try to change that. 297 00:16:45,560 --> 00:16:49,200 Speaker 2: They do things like modify tendons to try to make 298 00:16:49,320 --> 00:16:53,960 Speaker 2: muscles less stiff, or they transfer muscles to make them 299 00:16:53,960 --> 00:16:56,960 Speaker 2: do a new thing. But one of the big tricky 300 00:16:57,000 --> 00:17:00,240 Speaker 2: parts is that oftentimes some of those muscles are very week. 301 00:17:00,920 --> 00:17:03,600 Speaker 2: So if they choose the wrong muscle, then they'll make 302 00:17:03,600 --> 00:17:07,359 Speaker 2: a weak muscle even weaker. Huh, And that's catastrophic. So 303 00:17:07,400 --> 00:17:10,280 Speaker 2: it's a very fine line that a surgeon has to 304 00:17:10,280 --> 00:17:11,639 Speaker 2: figure out and they have to go in. You know, 305 00:17:11,720 --> 00:17:14,399 Speaker 2: we just talked about there's thirty five muscle muscles in 306 00:17:14,440 --> 00:17:16,359 Speaker 2: each leg, So which of the muscles are the ones 307 00:17:16,400 --> 00:17:19,480 Speaker 2: that should be operated on and which ones should be avoided? 308 00:17:20,200 --> 00:17:25,320 Speaker 2: And so what my collaborator, a guy named doctor Mark Abel, 309 00:17:26,119 --> 00:17:30,440 Speaker 2: fantastic surgeon. He was telling me, Yeah, like, it's very hard, 310 00:17:30,480 --> 00:17:32,480 Speaker 2: and I don't he didn't have a way to see that. 311 00:17:32,720 --> 00:17:34,840 Speaker 2: All that he could do is look from the outside. 312 00:17:35,280 --> 00:17:37,960 Speaker 2: No technology could give him the information he needed to 313 00:17:38,080 --> 00:17:40,840 Speaker 2: figure out which muscles he should focus on and which 314 00:17:40,840 --> 00:17:41,240 Speaker 2: ones to. 315 00:17:41,200 --> 00:17:44,240 Speaker 1: Avoid, because it's not obvious by looking what's a strong 316 00:17:44,320 --> 00:17:47,240 Speaker 1: muscle and what's a weak one. Yeah, I guess that's 317 00:17:47,240 --> 00:17:52,920 Speaker 1: surprising to me on some level, Like I don't I've 318 00:17:52,960 --> 00:17:55,600 Speaker 1: never thought about it. But naively, I would think you 319 00:17:55,680 --> 00:17:59,080 Speaker 1: could look at a muscle and say it looks strong 320 00:17:59,160 --> 00:17:59,720 Speaker 1: or it looks weak. 321 00:17:59,760 --> 00:18:02,200 Speaker 2: Not right, because you just see it from the surface. 322 00:18:02,320 --> 00:18:06,000 Speaker 2: You don't see it on the inside. And the other challenges. 323 00:18:06,640 --> 00:18:10,879 Speaker 2: For every joint, there's many muscles. So like we just 324 00:18:10,880 --> 00:18:13,439 Speaker 2: said that quadrceps has four muscles on the back of 325 00:18:13,440 --> 00:18:18,040 Speaker 2: the leg, ham strings, there's three hamstrings muscles. There's other 326 00:18:18,119 --> 00:18:21,480 Speaker 2: muscles that are in the thigh. So you're just seeing 327 00:18:21,520 --> 00:18:24,520 Speaker 2: what's an impairment and the overall movement, let's say of 328 00:18:24,560 --> 00:18:26,960 Speaker 2: a joint. But then there could be many muscles or 329 00:18:27,000 --> 00:18:29,119 Speaker 2: combinations of muscles that are leading to that, and you 330 00:18:29,160 --> 00:18:32,679 Speaker 2: don't know when you look from the outside. Our body 331 00:18:32,720 --> 00:18:34,640 Speaker 2: is designed that way actually to be somewhat we call 332 00:18:34,640 --> 00:18:37,440 Speaker 2: it redundant. We have more muscles than we need probably, 333 00:18:37,560 --> 00:18:41,560 Speaker 2: but if you think about imbalances, then any one of 334 00:18:41,600 --> 00:18:43,879 Speaker 2: those muscles could create it create some trouble. 335 00:18:44,119 --> 00:18:48,320 Speaker 1: So, okay, so the surgeon describes this problem he's having, 336 00:18:48,880 --> 00:18:50,520 Speaker 1: and then and then what do you do? 337 00:18:50,840 --> 00:18:53,359 Speaker 2: So then I was thinking, well, you know, that's the 338 00:18:53,400 --> 00:18:57,200 Speaker 2: information that we generate all the time. When we're doing 339 00:18:57,240 --> 00:19:03,240 Speaker 2: our research. We take MRIs, quantify, we identify muscles, we 340 00:19:03,359 --> 00:19:06,240 Speaker 2: create three dimensional models of the muscles, we figure out 341 00:19:06,240 --> 00:19:10,719 Speaker 2: how they're working from that. But I was struck by 342 00:19:10,760 --> 00:19:12,560 Speaker 2: the fact that none of that was something that a 343 00:19:12,600 --> 00:19:16,320 Speaker 2: clinician could use, despite the fact that MRI is obviously 344 00:19:16,440 --> 00:19:20,399 Speaker 2: ubiquitous in healthcare. Right, yeah, you can't go to a 345 00:19:20,400 --> 00:19:23,639 Speaker 2: hospital without finding multiple MRIs, but there's no way to 346 00:19:23,760 --> 00:19:26,000 Speaker 2: use those MRIs and the way that I was using 347 00:19:26,000 --> 00:19:30,200 Speaker 2: them for my research, And I thought, well, that's too bad, 348 00:19:30,280 --> 00:19:33,840 Speaker 2: because this would be very useful to the surgeon in 349 00:19:33,920 --> 00:19:38,560 Speaker 2: figuring out how to treat these patients. So that was 350 00:19:38,640 --> 00:19:41,439 Speaker 2: one light bulb at the beginning. So a lot of 351 00:19:41,480 --> 00:19:43,439 Speaker 2: it was figuring out how to take something that we 352 00:19:43,600 --> 00:19:48,240 Speaker 2: use in research for very specific, targeted basic science questions 353 00:19:48,560 --> 00:19:51,560 Speaker 2: and turn it into something that is useful clinically. 354 00:19:52,040 --> 00:19:55,080 Speaker 1: And again this is like my ignorance, Like I might 355 00:19:55,119 --> 00:19:58,119 Speaker 1: have thought, well, you could just do an MRI and 356 00:19:58,200 --> 00:20:02,280 Speaker 1: see how big or not big the muscles are, and 357 00:20:02,359 --> 00:20:04,920 Speaker 1: infer from how big or not big the muscles are, 358 00:20:05,800 --> 00:20:10,480 Speaker 1: how strong or not strong they are. And that sounds straightforward, 359 00:20:11,920 --> 00:20:14,960 Speaker 1: but clearly it's not, like, why is it harder than that? 360 00:20:16,320 --> 00:20:20,040 Speaker 2: So a couple of reasons. One is going taking the 361 00:20:20,160 --> 00:20:23,119 Speaker 2: MRI pictures and figuring out how big the muscles is 362 00:20:23,160 --> 00:20:27,800 Speaker 2: a very challenging problem. So in order to accurately get 363 00:20:27,840 --> 00:20:31,520 Speaker 2: how big the muscles are, you have to essentially generate 364 00:20:31,560 --> 00:20:33,840 Speaker 2: its shape in three dimensions, so you have to get 365 00:20:33,840 --> 00:20:37,000 Speaker 2: the whole length of the muscle, and so you do 366 00:20:37,119 --> 00:20:40,800 Speaker 2: that off of multiple MRI pictures. So the MRI essentially 367 00:20:40,840 --> 00:20:45,520 Speaker 2: kind of takes pictures through the body at multiple different 368 00:20:45,680 --> 00:20:49,960 Speaker 2: slices we call them, going from you know, the abdomen 369 00:20:50,000 --> 00:20:52,600 Speaker 2: all the way down to the feet, sort of going 370 00:20:52,840 --> 00:20:55,840 Speaker 2: cross sectionally we call it. And so we usually have 371 00:20:56,240 --> 00:21:00,439 Speaker 2: over two hundred images. So in each image you have 372 00:21:00,480 --> 00:21:03,480 Speaker 2: to find each muscle, and so for any given image 373 00:21:03,520 --> 00:21:08,800 Speaker 2: there's probably at least fifteen muscles or more. 374 00:21:09,359 --> 00:21:11,439 Speaker 1: So it wasn't like you could just push the like 375 00:21:11,840 --> 00:21:14,440 Speaker 1: show me the muscles button on the MRI and it 376 00:21:14,440 --> 00:21:15,359 Speaker 1: would show you the muscles. 377 00:21:15,359 --> 00:21:15,399 Speaker 2: Like. 378 00:21:15,440 --> 00:21:18,440 Speaker 1: Nobody had done that, and there was no obvious way 379 00:21:18,480 --> 00:21:21,560 Speaker 1: to do it, certainly not for a surgeon. Ordering a 380 00:21:21,920 --> 00:21:23,880 Speaker 1: standard MRON just didn't exist. 381 00:21:24,240 --> 00:21:25,159 Speaker 2: It did not exist. 382 00:21:25,560 --> 00:21:30,840 Speaker 1: So okay, so you realize this, what happens? How do 383 00:21:30,840 --> 00:21:31,520 Speaker 1: you make it happen? 384 00:21:32,280 --> 00:21:34,720 Speaker 2: So, you know, one of our first tasks was to 385 00:21:34,760 --> 00:21:38,480 Speaker 2: figure out how to get many muscles. So one of 386 00:21:38,520 --> 00:21:41,320 Speaker 2: the things that we had done on the research side 387 00:21:41,400 --> 00:21:43,920 Speaker 2: is really focus on a couple of muscles, but I 388 00:21:44,000 --> 00:21:46,600 Speaker 2: knew for this application that wasn't going to work. We 389 00:21:46,720 --> 00:21:49,840 Speaker 2: have to be able to identify any muscle. That was 390 00:21:49,840 --> 00:21:52,600 Speaker 2: really the problem is that like you don't know which 391 00:21:52,600 --> 00:21:54,399 Speaker 2: one's the problem, so you don't know which one to 392 00:21:54,440 --> 00:21:56,040 Speaker 2: look at. So you got to look at all of them. 393 00:21:56,520 --> 00:21:59,000 Speaker 2: And then the next task was to figure out all 394 00:21:59,000 --> 00:22:01,840 Speaker 2: those muscles and figure out a process to go from 395 00:22:01,880 --> 00:22:04,920 Speaker 2: the you know, identifying each and every muscle and each 396 00:22:04,960 --> 00:22:08,359 Speaker 2: and every image. So it's called developing an atlas. 397 00:22:08,560 --> 00:22:10,840 Speaker 1: And is that an AI problem? 398 00:22:12,080 --> 00:22:14,320 Speaker 2: So now we have an AI, and it's the type 399 00:22:14,320 --> 00:22:18,640 Speaker 2: of AI as supervised learning, where they can you essentially 400 00:22:18,800 --> 00:22:21,520 Speaker 2: train the computer to do what the person would do. 401 00:22:22,000 --> 00:22:23,879 Speaker 2: But in order to do that, you need to do 402 00:22:23,960 --> 00:22:28,680 Speaker 2: what the person would do first. And so we did 403 00:22:28,720 --> 00:22:32,920 Speaker 2: that all manually at first, in order to generate one 404 00:22:32,960 --> 00:22:36,159 Speaker 2: of these reports. At first, it took us about fifty 405 00:22:36,200 --> 00:22:38,800 Speaker 2: hours per person. 406 00:22:39,320 --> 00:22:42,920 Speaker 1: Just going through image after image after image and saying 407 00:22:42,920 --> 00:22:45,720 Speaker 1: this is this muscle, that is that muscle exactly. 408 00:22:45,960 --> 00:22:48,560 Speaker 2: So we needed to develop that. But the other piece 409 00:22:48,600 --> 00:22:51,840 Speaker 2: we needed is this back to this normative database I 410 00:22:51,880 --> 00:22:55,920 Speaker 2: talked about, because if I just told you how big 411 00:22:55,960 --> 00:22:58,959 Speaker 2: your muscle is in milli leaders in volume, what are 412 00:22:58,960 --> 00:23:00,000 Speaker 2: you going to do with that information? 413 00:23:00,600 --> 00:23:03,720 Speaker 1: Like, oh, great, and nobody knew And it's interesting. It's 414 00:23:03,720 --> 00:23:05,800 Speaker 1: one of those things you always think, oh, surely there's 415 00:23:05,800 --> 00:23:08,000 Speaker 1: some data in the world that everybody knows X. But 416 00:23:08,040 --> 00:23:10,600 Speaker 1: so you're saying, nobody knew what was the kind of 417 00:23:10,960 --> 00:23:15,399 Speaker 1: medium size of a particular quadricept for whatever, a healthy 418 00:23:15,440 --> 00:23:17,520 Speaker 1: twelve year old boy or whatever. Nobody knew that at 419 00:23:17,600 --> 00:23:18,960 Speaker 1: that time. Well, all of. 420 00:23:18,920 --> 00:23:22,879 Speaker 2: The information up until then, for the most part, was 421 00:23:22,920 --> 00:23:24,680 Speaker 2: based on dissecting cadavers. 422 00:23:24,960 --> 00:23:25,320 Speaker 1: Uh huh. 423 00:23:25,760 --> 00:23:32,520 Speaker 2: Was based on taking cadavers and dissecting muscles, weighing the muscles. 424 00:23:33,040 --> 00:23:35,080 Speaker 2: And you know, one of the big challenges with that 425 00:23:35,240 --> 00:23:40,399 Speaker 2: is usually cadavers are older adults, and so they're not 426 00:23:40,480 --> 00:23:46,320 Speaker 2: really representative of a younger, healthy population. And I will 427 00:23:46,320 --> 00:23:48,600 Speaker 2: tell you at that time that that was a lot 428 00:23:48,640 --> 00:23:50,800 Speaker 2: of work and we had I had people saying, like, 429 00:23:50,840 --> 00:23:54,080 Speaker 2: why are you doing that. Uh, like that seems like 430 00:23:54,119 --> 00:23:58,160 Speaker 2: a waste of time. That's crazy. You know, I had 431 00:23:58,160 --> 00:24:00,760 Speaker 2: this vision and I trusted that it was going to 432 00:24:01,040 --> 00:24:04,200 Speaker 2: turn into something at least useful to the research community, 433 00:24:04,200 --> 00:24:06,800 Speaker 2: and you know, I'm thankful that we stuck with it. 434 00:24:08,600 --> 00:24:10,840 Speaker 1: There's lots more to come on the show, including but 435 00:24:11,040 --> 00:24:14,199 Speaker 1: not limited to the work Sylvia and her colleagues are 436 00:24:14,200 --> 00:24:18,840 Speaker 1: doing with major league pitchers, college football players, and patients 437 00:24:18,920 --> 00:24:33,760 Speaker 1: with degenerative muscle disease. Sylvia and her colleagues trained an 438 00:24:33,800 --> 00:24:36,840 Speaker 1: AI model to do what had previously taken a human 439 00:24:37,320 --> 00:24:41,560 Speaker 1: fifty hours for every person who got scanned, and they 440 00:24:41,680 --> 00:24:45,800 Speaker 1: expanded from working with patients with cerebral palsy to working 441 00:24:45,840 --> 00:24:50,640 Speaker 1: with elite athletes. Today, their clients include not just Olympic athletes, 442 00:24:51,000 --> 00:24:54,520 Speaker 1: but teams in the NBA and the Premier League. Also, 443 00:24:54,720 --> 00:24:57,040 Speaker 1: she told me they're working on a project with Major 444 00:24:57,119 --> 00:24:57,800 Speaker 1: League Baseball. 445 00:24:58,600 --> 00:25:02,040 Speaker 2: Yeah, so we're working with the MLB studying pictures and 446 00:25:02,080 --> 00:25:05,800 Speaker 2: we're getting essentially a normative database, whole body scan of pictures. 447 00:25:05,840 --> 00:25:09,159 Speaker 1: And is that partly because like pitchers mess up their 448 00:25:09,280 --> 00:25:13,040 Speaker 1: arms so badly? Is that kind of the motivation there, Yes. 449 00:25:12,920 --> 00:25:17,840 Speaker 2: There's definitely a lot of issue with with injury and surgery. 450 00:25:17,960 --> 00:25:21,800 Speaker 2: And so the idea here is that by taking these scans, 451 00:25:21,840 --> 00:25:25,280 Speaker 2: we can really figure out where there might be weaknesses 452 00:25:25,440 --> 00:25:29,680 Speaker 2: and sort of potential areas for mitigating the injuries. 453 00:25:30,080 --> 00:25:34,159 Speaker 1: So when you do work for a whole team, like 454 00:25:34,240 --> 00:25:38,119 Speaker 1: say the Bulls, you know, a basketball team, an NBA team, Like, 455 00:25:38,359 --> 00:25:40,679 Speaker 1: what's the nature of that of that work? What do 456 00:25:40,720 --> 00:25:41,960 Speaker 1: you do for a team like that? 457 00:25:42,720 --> 00:25:46,359 Speaker 2: Yeah, we're they will do a baseline of the whole 458 00:25:46,359 --> 00:25:46,960 Speaker 2: team and. 459 00:25:47,320 --> 00:25:52,639 Speaker 1: They basically tailor the athletes training presumably strength training in 460 00:25:52,680 --> 00:25:57,600 Speaker 1: particular on a muscle by muscle basis, based on the 461 00:25:57,640 --> 00:25:58,880 Speaker 1: reports that you're sending them. 462 00:25:59,040 --> 00:25:59,399 Speaker 2: Correct. 463 00:25:59,480 --> 00:26:03,840 Speaker 1: Yeah, yeah, And I mean you can imagine like better 464 00:26:03,880 --> 00:26:09,880 Speaker 1: performance being one outcome. Reduced risk of injury seems plausible, right, 465 00:26:10,080 --> 00:26:13,720 Speaker 1: Like it seems obvious that like a big asymmetry could 466 00:26:14,119 --> 00:26:16,680 Speaker 1: make you more likely to be injured. I mean, are 467 00:26:16,720 --> 00:26:19,199 Speaker 1: you at a point now where you can predict the 468 00:26:19,280 --> 00:26:22,640 Speaker 1: risk of injury? 469 00:26:22,760 --> 00:26:24,159 Speaker 2: That's like a whole can of worms. 470 00:26:24,560 --> 00:26:28,760 Speaker 1: I won't say, I mean, is that interesting to you? 471 00:26:28,920 --> 00:26:30,800 Speaker 1: Or is that like too much? Or yeah? 472 00:26:30,880 --> 00:26:33,280 Speaker 2: No, no, no, this is something we think about a lot, 473 00:26:34,000 --> 00:26:38,480 Speaker 2: and let me I want to so first, I'll tell 474 00:26:38,480 --> 00:26:40,359 Speaker 2: you why it's a can of worms. Yeah yeah, And 475 00:26:40,359 --> 00:26:41,960 Speaker 2: I'll tell you what project. 476 00:26:42,200 --> 00:26:43,760 Speaker 1: Tell me about the can. We'll look at it from 477 00:26:43,760 --> 00:26:44,359 Speaker 1: the outside of. 478 00:26:44,400 --> 00:26:46,199 Speaker 2: It, so from the cannon. Like, there's a lot of 479 00:26:46,240 --> 00:26:49,840 Speaker 2: technologies out there that will say that they're predicting injury risk. 480 00:26:50,400 --> 00:26:53,920 Speaker 2: They'll give you numbers, and they're just not based on anything, 481 00:26:55,960 --> 00:26:58,560 Speaker 2: and so I don't know, it's there's. 482 00:26:58,359 --> 00:27:02,320 Speaker 1: A lot of the yeah. 483 00:27:02,160 --> 00:27:05,280 Speaker 2: Yeah, yeah, yeah. So that's not what we're about, Like, 484 00:27:05,480 --> 00:27:10,080 Speaker 2: we're about like providing actual things that matter. And so 485 00:27:10,680 --> 00:27:14,920 Speaker 2: the question is like can you do do these muscle scans? 486 00:27:16,119 --> 00:27:20,919 Speaker 2: Do they correlate with injury likelihood in some way? And 487 00:27:20,960 --> 00:27:24,280 Speaker 2: so we actually have a project to address that very question. 488 00:27:24,640 --> 00:27:28,600 Speaker 2: It's actually funded by the NFL. Uh. We're actually in 489 00:27:28,640 --> 00:27:34,000 Speaker 2: that project. We're working with college teams called college football 490 00:27:34,000 --> 00:27:38,880 Speaker 2: teams baselining entire rosters at the beginning of the season 491 00:27:39,560 --> 00:27:43,800 Speaker 2: and then tracking hamstring injuries and then if a if 492 00:27:43,800 --> 00:27:46,240 Speaker 2: an athlete gets injured, they come back for a scan 493 00:27:46,600 --> 00:27:49,399 Speaker 2: at the time of injury and then return to sport. 494 00:27:50,160 --> 00:27:52,720 Speaker 2: And so one of our questions is based on the 495 00:27:53,240 --> 00:27:57,600 Speaker 2: baseline scan, can we predict who's more likely to get 496 00:27:57,640 --> 00:28:01,400 Speaker 2: an injury, an initial injury index injury, and then then 497 00:28:01,440 --> 00:28:04,040 Speaker 2: the secondary question is can we predict who will be 498 00:28:04,160 --> 00:28:07,639 Speaker 2: re injured? I was saying that we often pairt people 499 00:28:07,680 --> 00:28:09,920 Speaker 2: pair it with other things. In this project, we're always 500 00:28:09,960 --> 00:28:13,840 Speaker 2: also doing that. Each athlete is getting an assessment of 501 00:28:13,880 --> 00:28:17,199 Speaker 2: their sprint mechanics, so kind of the biomechanics of how 502 00:28:17,240 --> 00:28:22,119 Speaker 2: they run, and then also assessment of their strength of 503 00:28:22,160 --> 00:28:25,639 Speaker 2: their hamstring muscles, so like kind of measured strength. Obviously 504 00:28:25,720 --> 00:28:27,960 Speaker 2: you can't do that when they have an injury, of course, 505 00:28:28,080 --> 00:28:30,240 Speaker 2: but you can you know when they're healthy. 506 00:28:30,400 --> 00:28:33,840 Speaker 1: So that biomechanics piece seems like something that has been 507 00:28:33,840 --> 00:28:39,760 Speaker 1: developing in parallel with your work, also driven by computer vision, right, that, 508 00:28:40,000 --> 00:28:47,800 Speaker 1: like markerless motion captures, seems like a big world that 509 00:28:47,800 --> 00:28:52,200 Speaker 1: that overlaps with your world some Yeah, So tell me 510 00:28:52,240 --> 00:28:56,720 Speaker 1: about your work with female athletes versus male athletes and 511 00:28:55,960 --> 00:28:58,440 Speaker 1: how that plays a role. 512 00:28:59,200 --> 00:29:02,080 Speaker 2: Yeah, I will say probably the biggest thing that we've 513 00:29:02,680 --> 00:29:07,360 Speaker 2: been focused on is making sure that our data addresses that. 514 00:29:07,600 --> 00:29:13,360 Speaker 2: So our normative database is uh separated by sex, so 515 00:29:13,760 --> 00:29:18,320 Speaker 2: and it is different because like women aren't small men, right, 516 00:29:19,000 --> 00:29:22,800 Speaker 2: So it's important that we have that basis to compare 517 00:29:22,880 --> 00:29:26,960 Speaker 2: that's like for women and not comparing to some average 518 00:29:27,080 --> 00:29:30,760 Speaker 2: or primarily male data sets. So that's that's one huge 519 00:29:30,800 --> 00:29:34,360 Speaker 2: important thing is that it's it's compared to the normative 520 00:29:34,400 --> 00:29:41,000 Speaker 2: values for the female population. And then in terms of 521 00:29:41,080 --> 00:29:43,680 Speaker 2: like working with the female athletes, I think, you know, 522 00:29:43,760 --> 00:29:46,280 Speaker 2: one of the big ones is is really just ability 523 00:29:46,360 --> 00:29:53,720 Speaker 2: to personalize and provide this like really accurate detailed assessment 524 00:29:53,800 --> 00:29:56,960 Speaker 2: of their of their bodies. And you know, a lot 525 00:29:57,000 --> 00:30:00,840 Speaker 2: of the you know, knowledge about like appropriate body composition 526 00:30:02,080 --> 00:30:06,120 Speaker 2: historically has been based on studies and men, and but 527 00:30:06,160 --> 00:30:09,320 Speaker 2: then we're applying them to women and making us feel 528 00:30:09,320 --> 00:30:14,360 Speaker 2: really bad about ourselves. So really motivated to move away 529 00:30:14,400 --> 00:30:18,520 Speaker 2: from that and sort of acknowledge the muscular physiology and 530 00:30:18,600 --> 00:30:22,560 Speaker 2: anatomy of the female and also the female athlete to 531 00:30:22,600 --> 00:30:26,800 Speaker 2: really understand understand that. You know, I think one thing 532 00:30:26,840 --> 00:30:29,800 Speaker 2: obviously that we've seen is, you know, acl injuries are 533 00:30:30,120 --> 00:30:35,520 Speaker 2: more common in women than men, and examining how these 534 00:30:35,640 --> 00:30:39,280 Speaker 2: like recovery profiles look on how they differ between men 535 00:30:39,320 --> 00:30:42,720 Speaker 2: and women. That's something that we're observing and seeing how 536 00:30:42,760 --> 00:30:46,720 Speaker 2: those things shake out. But we're motivated by really providing 537 00:30:46,760 --> 00:30:48,600 Speaker 2: that information that's specific to women. 538 00:30:49,720 --> 00:30:53,560 Speaker 1: So what are some of the non sports things you're 539 00:30:53,600 --> 00:30:55,480 Speaker 1: working on, things you're trying to figure out. 540 00:30:55,760 --> 00:31:00,760 Speaker 2: Yeah, I mean, one that's I'm really interested in is 541 00:31:01,560 --> 00:31:04,720 Speaker 2: this area that we're applying to in clinical trials for 542 00:31:04,840 --> 00:31:09,680 Speaker 2: muscle disease. So we've been working in a specific muscle 543 00:31:09,720 --> 00:31:15,920 Speaker 2: disease called fascioscapulo humoral muscular district f SHD, which is 544 00:31:15,960 --> 00:31:21,000 Speaker 2: a slowly progressing muscle disease genetic and basis, and so 545 00:31:22,280 --> 00:31:26,760 Speaker 2: eventually people with f SHD need a wheelchair. Just life 546 00:31:26,800 --> 00:31:30,480 Speaker 2: is very difficult, and so it's pretty devastating. But the 547 00:31:30,520 --> 00:31:33,520 Speaker 2: other exciting thing is there are some new treatments out there, 548 00:31:33,680 --> 00:31:37,400 Speaker 2: some in particular gene therapies coming online. And now the 549 00:31:37,480 --> 00:31:41,560 Speaker 2: challenges do they work because the problem is in these 550 00:31:41,600 --> 00:31:45,200 Speaker 2: diseases because they're pretty slowly progressing. If you want to 551 00:31:45,240 --> 00:31:50,160 Speaker 2: see if a drug is helping somebody, it's very hard 552 00:31:50,200 --> 00:31:52,080 Speaker 2: to see that in a slowly progressing. 553 00:31:51,840 --> 00:31:55,120 Speaker 1: Right and the clinical manifestations are hard to pick up. 554 00:31:55,760 --> 00:32:00,640 Speaker 1: If it makes your muscles shrink more slowly, it's going 555 00:32:00,720 --> 00:32:01,360 Speaker 1: to be hard to see. 556 00:32:01,520 --> 00:32:04,600 Speaker 2: It's very hard to see, especially from like rudimentary measures. 557 00:32:04,880 --> 00:32:07,840 Speaker 2: But with the MRIs. We've been able to provide this 558 00:32:07,960 --> 00:32:11,840 Speaker 2: really detailed insight about the disease date of each muscle 559 00:32:11,880 --> 00:32:15,280 Speaker 2: and how it's progressing over time. And so you know, 560 00:32:15,320 --> 00:32:17,400 Speaker 2: one of our goals is to really lean in on 561 00:32:17,520 --> 00:32:21,720 Speaker 2: IS and help figure out exactly how people should look 562 00:32:21,720 --> 00:32:24,240 Speaker 2: at all this data and figure out if a drug 563 00:32:24,320 --> 00:32:27,960 Speaker 2: is working or not. It's really profoundly important because without that, 564 00:32:28,320 --> 00:32:31,040 Speaker 2: these these clinical trials just won't move forward. 565 00:32:32,720 --> 00:32:37,120 Speaker 1: What else are you sort of still trying to figure out? 566 00:32:37,680 --> 00:32:42,440 Speaker 2: So we talked about predicting injury but having all the 567 00:32:42,520 --> 00:32:45,520 Speaker 2: data needed to show like if you do if your 568 00:32:45,520 --> 00:32:47,680 Speaker 2: scan looks like this, and if you do this, you 569 00:32:47,720 --> 00:32:50,000 Speaker 2: will be able to improve your jump high by height. 570 00:32:50,080 --> 00:32:52,800 Speaker 1: By that, yeah, you'll be able to throw a fastball 571 00:32:53,320 --> 00:32:57,200 Speaker 1: two miles an hour faster. Like that would be wildly valuable. 572 00:32:57,440 --> 00:32:59,640 Speaker 2: That would be very and we do have data in 573 00:32:59,680 --> 00:33:04,720 Speaker 2: our search. We were able to show that these muscle 574 00:33:04,840 --> 00:33:09,080 Speaker 2: scores correlate with performance metrics such as jump, hide, and speed. 575 00:33:09,400 --> 00:33:12,240 Speaker 2: So we for sure see that The question is then 576 00:33:12,280 --> 00:33:15,720 Speaker 2: the spin on like observing how how that plays out, 577 00:33:15,840 --> 00:33:20,680 Speaker 2: Like if you then strengthen the appropriate muscles, how how 578 00:33:20,760 --> 00:33:22,840 Speaker 2: much faster do you get? And you just need more 579 00:33:22,920 --> 00:33:26,080 Speaker 2: and more data to really like to go after that. 580 00:33:26,120 --> 00:33:29,040 Speaker 2: But that's one thing that I'm fascinated by. One of 581 00:33:29,080 --> 00:33:32,280 Speaker 2: the other interesting ones, Can I go off on a tangent. 582 00:33:32,040 --> 00:33:35,320 Speaker 1: Anything you want? 583 00:33:35,320 --> 00:33:38,800 Speaker 2: One of our research partners that's interested in how muscles 584 00:33:38,840 --> 00:33:42,400 Speaker 2: adapt to strength training and different interventions and what influences 585 00:33:42,440 --> 00:33:46,040 Speaker 2: that had a really interesting finding that I think is 586 00:33:46,120 --> 00:33:50,080 Speaker 2: quite profound but also obvious. So everybody if there, if 587 00:33:50,120 --> 00:33:55,040 Speaker 2: they're targeted training their quadrceps and hamstrings, those muscles got bigger, 588 00:33:56,040 --> 00:33:59,520 Speaker 2: that makes sense, but in a fair number of the people, 589 00:33:59,680 --> 00:34:04,560 Speaker 2: some muscles got smaller. And then you know, he had 590 00:34:04,640 --> 00:34:10,719 Speaker 2: done some controlling and in documentation of nutrition intake, and 591 00:34:10,760 --> 00:34:15,480 Speaker 2: he found that people that had higher caloric and protein 592 00:34:15,520 --> 00:34:18,760 Speaker 2: intake had less of that effect. 593 00:34:19,120 --> 00:34:21,319 Speaker 1: So all that, all the Jim bros. Telling you to 594 00:34:21,360 --> 00:34:25,120 Speaker 1: eat a lot of protein are validated by this guy's study. 595 00:34:25,239 --> 00:34:27,560 Speaker 2: Yeah, yeah, but they're not. It's not necessarily to make 596 00:34:27,600 --> 00:34:29,719 Speaker 2: that muscle that you're working bigger so you don't lose 597 00:34:29,760 --> 00:34:30,400 Speaker 2: the other muscles. 598 00:34:30,480 --> 00:34:35,400 Speaker 1: Uh huh. That's a good one. And was he using 599 00:34:35,440 --> 00:34:36,960 Speaker 1: your scans to figure out that? 600 00:34:37,280 --> 00:34:39,400 Speaker 2: Yeah? Yeah, he was using our scans and and the 601 00:34:39,480 --> 00:34:42,560 Speaker 2: thing that was cool. Is that Normally in research you 602 00:34:42,600 --> 00:34:45,000 Speaker 2: wouldn't bother looking at those other muscles. You would just 603 00:34:45,000 --> 00:34:47,040 Speaker 2: look at the ones that were targeted, right, because those 604 00:34:47,080 --> 00:34:49,000 Speaker 2: are the ones that you just think about. But by 605 00:34:49,040 --> 00:34:52,480 Speaker 2: getting the entire extent of the of all the muscles, 606 00:34:52,920 --> 00:34:56,200 Speaker 2: you see these impacts that you wouldn't necessarily. 607 00:34:55,680 --> 00:34:57,880 Speaker 1: Have known, Like he wasn't even looking for it. 608 00:34:58,160 --> 00:35:02,280 Speaker 2: Yeah, it's it's quite profound because somebody's strength training recovering 609 00:35:02,320 --> 00:35:06,360 Speaker 2: from an injury. That really means like that, the nutritional 610 00:35:06,400 --> 00:35:10,960 Speaker 2: elements important because you could be strengthening some muscles but 611 00:35:11,000 --> 00:35:13,520 Speaker 2: weakening others if you're not. If you're not, you know, 612 00:35:13,520 --> 00:35:14,600 Speaker 2: playing your cards right there. 613 00:35:18,000 --> 00:35:20,040 Speaker 1: We'll be back in a minute with the lightning round. 614 00:35:21,080 --> 00:35:32,640 Speaker 1: M h, I want to finish with the lightning round. 615 00:35:33,000 --> 00:35:34,959 Speaker 1: I won't take too long. It'll be fun. 616 00:35:35,320 --> 00:35:38,200 Speaker 2: Okay, what I don't know what that is? 617 00:35:38,440 --> 00:35:40,360 Speaker 1: Well, you'll find out right now. 618 00:35:41,760 --> 00:35:42,840 Speaker 2: Does this have to be fast? 619 00:35:44,800 --> 00:35:47,200 Speaker 1: I could call it the random round? Okay, I like 620 00:35:47,239 --> 00:35:51,560 Speaker 1: that random. Have you scanned yourself? 621 00:35:52,200 --> 00:35:57,399 Speaker 2: Yes? Well times? Oh of course that's what kind of thing. 622 00:35:58,280 --> 00:36:00,279 Speaker 2: Probably I've been in an MRI machine, I don't know, 623 00:36:00,360 --> 00:36:01,360 Speaker 2: maybe a hundred times. 624 00:36:01,400 --> 00:36:04,359 Speaker 1: Like it's not radiation. Right, it's not like an X ray. 625 00:36:04,440 --> 00:36:05,640 Speaker 1: You could do it every day. 626 00:36:05,520 --> 00:36:06,239 Speaker 2: As much as you want. 627 00:36:06,320 --> 00:36:08,399 Speaker 1: Yeah, what'd you learn? 628 00:36:09,840 --> 00:36:13,360 Speaker 2: So I actually used it? You know, I've learned lots 629 00:36:13,440 --> 00:36:15,800 Speaker 2: over the years, but I will tell you one anecdote. 630 00:36:15,840 --> 00:36:21,360 Speaker 2: I have a hip replacement. I have a genetic condition 631 00:36:21,440 --> 00:36:26,240 Speaker 2: that leads to early arthritis. And so I was before 632 00:36:26,320 --> 00:36:28,799 Speaker 2: I got my hip replacement, I got a scan. I 633 00:36:28,920 --> 00:36:31,600 Speaker 2: knew I was getting weak, but holy cow, was I 634 00:36:31,680 --> 00:36:34,960 Speaker 2: really weak on that side. What was profound was how 635 00:36:36,040 --> 00:36:41,040 Speaker 2: weak my hip flexers were very weak. And I you know, 636 00:36:41,080 --> 00:36:42,960 Speaker 2: I think a lot of times people talk about hip 637 00:36:42,960 --> 00:36:46,280 Speaker 2: flexures being tight, and that's kind of what I thought 638 00:36:46,360 --> 00:36:49,600 Speaker 2: was happening. I felt pain and I felt like I 639 00:36:49,680 --> 00:36:51,800 Speaker 2: was having a lot of tightness, but it was actually 640 00:36:52,320 --> 00:36:57,560 Speaker 2: weakness and they were like super small on both sides, 641 00:36:57,600 --> 00:36:59,680 Speaker 2: but really especially on the on the side that was 642 00:36:59,680 --> 00:37:03,400 Speaker 2: effect it. So that was one thing that I worked 643 00:37:03,440 --> 00:37:03,959 Speaker 2: on a lot. 644 00:37:04,360 --> 00:37:07,240 Speaker 1: Do that genetic condition you have, did that influence your 645 00:37:07,920 --> 00:37:10,400 Speaker 1: your work and all your decision to go into the field. 646 00:37:13,560 --> 00:37:17,080 Speaker 2: I mean it's like loosely maybe because my dad had 647 00:37:17,120 --> 00:37:21,360 Speaker 2: the same thing, which actually caused him to go blind. Oh, 648 00:37:21,480 --> 00:37:23,920 Speaker 2: it has like a multiple different issues, and so I 649 00:37:23,920 --> 00:37:26,640 Speaker 2: think that at an early age got me interested in 650 00:37:27,960 --> 00:37:35,000 Speaker 2: medicine and disabilities and like helping helping people, so that 651 00:37:35,000 --> 00:37:37,359 Speaker 2: that might be broadly speaking, I didn't and I knew 652 00:37:37,400 --> 00:37:39,680 Speaker 2: I had some eye problems. I didn't know the genetic thing. 653 00:37:40,200 --> 00:37:41,640 Speaker 2: We didn't discover that till later. 654 00:37:41,719 --> 00:37:47,239 Speaker 1: But huh. Interesting. What's the most underrated muscle in the 655 00:37:47,320 --> 00:37:47,960 Speaker 1: human body? 656 00:37:48,600 --> 00:37:56,120 Speaker 2: Hmm, that's a hard one. So I have a few 657 00:37:56,160 --> 00:37:57,040 Speaker 2: favorite muscles. 658 00:37:57,560 --> 00:37:58,760 Speaker 1: Okay, what's your favorite muscle? 659 00:37:58,920 --> 00:38:02,120 Speaker 2: Yeah, so the so muscle so as major it's it's 660 00:38:02,120 --> 00:38:04,839 Speaker 2: a hip flexer, okay, but it also it's really cool 661 00:38:05,239 --> 00:38:09,080 Speaker 2: it actually it's also a back lower back muscle, so 662 00:38:09,120 --> 00:38:12,800 Speaker 2: it attaches to the lumbar vertebrae. But then it also 663 00:38:13,560 --> 00:38:16,000 Speaker 2: crosses the front of your hip. It's really hard to 664 00:38:16,080 --> 00:38:18,600 Speaker 2: find because it's like really back deep in your hip. 665 00:38:19,160 --> 00:38:20,839 Speaker 2: It goes right over your. 666 00:38:20,640 --> 00:38:24,439 Speaker 1: Your in the middle of your body kind of yeah, 667 00:38:24,640 --> 00:38:25,280 Speaker 1: right in the middle. 668 00:38:25,440 --> 00:38:28,680 Speaker 2: It kind of connects everything, sort of connects your lower 669 00:38:28,719 --> 00:38:31,200 Speaker 2: extremity to the rest of your body in some ways. 670 00:38:31,719 --> 00:38:39,680 Speaker 1: Okay, last one, why do you hate astrophysicist Barbie? 671 00:38:40,320 --> 00:38:46,920 Speaker 2: I don't hate anything. I mean, well, it's too perfect. 672 00:38:47,200 --> 00:38:50,799 Speaker 2: It's kind of like this, you know idea that like, oh, 673 00:38:51,200 --> 00:38:53,880 Speaker 2: you know, you can, you can inspire girls to go 674 00:38:54,000 --> 00:38:57,080 Speaker 2: into science by showing them that Barbie does too. But 675 00:38:57,200 --> 00:39:00,600 Speaker 2: Barbie's like fictitious, so it kind of tells you that, 676 00:39:02,160 --> 00:39:07,480 Speaker 2: like it like promotes the idea of perfectionism in society 677 00:39:07,480 --> 00:39:10,799 Speaker 2: but definitely in girls. And you know, what we really 678 00:39:10,800 --> 00:39:13,560 Speaker 2: want to promote is almost the opposite of that is 679 00:39:13,600 --> 00:39:18,560 Speaker 2: taking risks and and not worrying about being perfect and 680 00:39:18,640 --> 00:39:22,920 Speaker 2: just doing something that matters to you. So yeah, I 681 00:39:22,960 --> 00:39:24,600 Speaker 2: don't know. I mean, I don't hate it. I had 682 00:39:24,600 --> 00:39:26,719 Speaker 2: Barbies when I was a kid, but I just it 683 00:39:26,840 --> 00:39:27,200 Speaker 2: kind of. 684 00:39:27,160 --> 00:39:32,680 Speaker 1: Like wrote you wrote a whole column about it. 685 00:39:34,760 --> 00:39:38,719 Speaker 2: That was I was very proud of that. Yeah, no, 686 00:39:38,840 --> 00:39:41,800 Speaker 2: and really that, you know, the astrophysicist part, honestly was 687 00:39:41,840 --> 00:39:45,400 Speaker 2: more of a hook. The article I had written already 688 00:39:45,520 --> 00:39:49,160 Speaker 2: before that Astrophysicist Barbie came to be. It was about 689 00:39:49,200 --> 00:39:55,000 Speaker 2: the issue of perfectionism and how that dissuades girls to 690 00:39:55,040 --> 00:39:57,640 Speaker 2: go into stem and research. 691 00:39:58,280 --> 00:40:01,879 Speaker 1: Well, a good hook is important, and you found. 692 00:40:03,200 --> 00:40:03,399 Speaker 2: Yeah. 693 00:40:07,280 --> 00:40:10,280 Speaker 1: Sylvia Bleimker is a professor at the University of Virginia 694 00:40:10,480 --> 00:40:14,799 Speaker 1: and the co founder of Springbok analytics. Next week on 695 00:40:14,840 --> 00:40:17,960 Speaker 1: What's Your Problem, I'll be talking to Jimmy Buffy. He 696 00:40:18,120 --> 00:40:21,840 Speaker 1: is using AI to bring the insights of biomechanics to 697 00:40:21,960 --> 00:40:26,400 Speaker 1: professional athletes. Jimmy told me that before the advent of AI, 698 00:40:26,760 --> 00:40:30,200 Speaker 1: when biomechanics experts tried to work with athletes, it could 699 00:40:30,239 --> 00:40:33,400 Speaker 1: be somewhat awkward. So You've got like a picture in 700 00:40:33,440 --> 00:40:36,719 Speaker 1: his underwear with a bunch of little metal balls, Take 701 00:40:36,840 --> 00:40:39,239 Speaker 1: them and they're like, just pitch like you always pitched right. 702 00:40:39,239 --> 00:40:40,920 Speaker 3: And the state of the art for tracking it was 703 00:40:41,000 --> 00:40:43,200 Speaker 3: an awful experience for the people you. 704 00:40:43,160 --> 00:40:43,839 Speaker 2: Were trying to track. 705 00:40:45,000 --> 00:40:45,920 Speaker 1: So what changes? 706 00:40:46,600 --> 00:40:50,840 Speaker 3: The big inflection point was computer vision, basically using artificial 707 00:40:50,880 --> 00:40:55,800 Speaker 3: intelligence to identify where those joints are in a camera image, 708 00:40:55,920 --> 00:40:59,640 Speaker 3: rather than needing to paste those reflective markers. 709 00:41:00,960 --> 00:41:04,400 Speaker 1: Today's show was produced by Gabriel Hunter Chang, edited by 710 00:41:04,480 --> 00:41:08,640 Speaker 1: Lydia Jean Kott, and engineered by Sarah Bugero. You can 711 00:41:08,640 --> 00:41:13,240 Speaker 1: email us at Problem at Pushkin dot FM. I'm Jacob Goldstein. 712 00:41:13,320 --> 00:41:15,640 Speaker 1: We'll be back next week with another episode of What's 713 00:41:15,680 --> 00:41:29,160 Speaker 1: Your Problem.