Latest / Automated with Brian Heater / Kathryn Zealand and Jessica Bath on Wearable Robotics, Parkinson’s, and the Future of Mobility

Transcript

Kathryn Zealand and Jessica Bath on Wearable Robotics, Parkinson’s, and the Future of Mobility · Automated with Brian Heater

• 0:00
  SPEAKER_03: Is there enough information in someone's gate to understand what movement they're doing before they do it?
• 0:05
  SPEAKER_01: It's like predictive text for walking.
• 0:08
  SPEAKER_03: Exactly.
• 0:08
  SPEAKER_03: My grandmother hit 90, one day she had a fall, lost the confidence to do that walk, and she just stopped doing it.
• 0:15
  SPEAKER_03: I came to view physical movement and the ability to kind of be active and be a part of your community as essential to kind of physical and mental health care.
• 0:22
  SPEAKER_03: There's just no good solution.
• 0:24
  SPEAKER_03: Part of the reason our first product mogo is actually more about hiking and aspiration, is we're trying to pitch it more like an e-bike.
• 0:30
  SPEAKER_03: They're for everyone and they can help anyone.
• 0:32
  SPEAKER_03: But it happens that the people who probably get the most benefit out of e-bikes tend to be a little bit older.
• 0:36
  SPEAKER_01: I really wanted to talk about the Parkinson's thing because this is something that's impersonal.
• 0:41
  SPEAKER_01: In my life, my my father was diagnosed with Parkinson's.
• 0:44
  SPEAKER_01: When looking at Parkinson's specifically versus other mobility issues, what are you targeting?
• 0:49
  SPEAKER_03: Three different things.
• 0:50
  SPEAKER_03: So one is just bucket number two is f bucket number three, which is which I should say is the hardest and the one that you know I might get in trouble for talking about because hello everyone.
• 1:14
  SPEAKER_01: My name is Brian Heater.
• 1:16
  SPEAKER_01: I am the managing editor at the Association for Advancing Automation.
• 1:20
  SPEAKER_01: We've got a special episode for you this week.
• 1:22
  SPEAKER_01: We're speaking with Catherine Zeeland of the Google X soft exoskeleton spinoff Skip.
• 1:28
  SPEAKER_01: Catherine and I get a bit into the company's work around Parkinson's disease, which unfortunately has become a very personal topic for me over the last several years.
• 1:38
  SPEAKER_01: Catherine was also kind enough to introduce us to Jessica Bath of UCSF.
• 1:44
  SPEAKER_01: She's been working with Skip and gives us some insights into the tech work being done in and around Parkinson's.
• 1:51
  SPEAKER_01: Thanks to everybody who has subscribed to the show recently.
• 1:54
  SPEAKER_01: Getting those numbers up is exactly what allows us to do the show every week.
• 1:57
  SPEAKER_01: So please like and subscribe if you haven't already.
• 2:00
  SPEAKER_01: In the meantime, please enjoy this conversation with Catherine Zeeland and Jessica Bath.
• 2:05
  SPEAKER_01: Vision and AI are changing what's possible in industrial automation, from quality inspection and predictive maintenance to robotics, logistics, medical devices, aerospace, and manufacturing.
• 2:17
  SPEAKER_01: At the Advanced Vision and AI conference, you'll hear from industry leaders, see real-world AI applications in action, explore the latest in machine vision, computer vision, sensors, edge computing, and automation technology.
• 2:29
  SPEAKER_01: You'll also be able to connect with the people who are shaping what comes next.
• 2:33
  SPEAKER_01: Whether you're building, buying, or implementing these systems, this is a chance to stay ahead of the curve and see how vision and AI are moving from promise to practical impact.
• 2:44
  SPEAKER_01: Learn more and make sure to buy your ticket early at the Advanced Vision and AI conference September 23rd and 24th in Santa Clara.
• 2:54
  SPEAKER_01: So do you keep up with all of the, I don't want to say black hole news, but do you keep up with all of the sort of latest like black hole related findings, related happenings?
• 3:06
  SPEAKER_03: A little bit.
• 3:07
  SPEAKER_03: Like I'm no longer on the cutting edge, but especially the area that I used to work in, every now and again I pop in and I'm like, has anyone solved my PhD problem yet?
• 3:14
  SPEAKER_03: Because I could I quit full disclosure before solving it.
• 3:17
  SPEAKER_03: And part of me is like a little bit glad that it hasn't yet been solved, you know, because then I feel, okay, I didn't like it, I didn't miss something obvious.
• 3:24
  SPEAKER_03: It wasn't a total waste of time.
• 3:26
  SPEAKER_03: Um yeah, it's it's fun.
• 3:29
  SPEAKER_01: Yeah, that was something fascinating that you told me the last time we spoke, is that your like entire PhD thesis was tied to basically a single equation, which is very, which is very like all in, right?
• 3:44
  SPEAKER_01: I mean, it's either it's either it's either solved or or or it isn't.
• 3:49
  SPEAKER_01: And is that I don't know, I guess is your ability to get that degree, does it hinge on your ability specifically to solve that equation, or can you still sort of get through the program and not have solved that?
• 4:02
  SPEAKER_03: I mean, I think in theoretical physics, you have to show that you've furthered science.
• 4:07
  SPEAKER_03: So you can make a case that, well, I explored many avenues and furthered our understanding of what's not possible in different approaches, but then those are those tend to be longer and harder PhDs committed to Facebook.
• 4:19
  SPEAKER_03: Whereas if you actually solve the problem, um then that tends to be a much clearer cut, easier uh PhD route.
• 4:27
  SPEAKER_03: Um, it's been a problem.
• 4:28
  SPEAKER_03: I've had friends who would spend a decade and not and not have much to show for it because nothing they tried worked.
• 4:35
  SPEAKER_01: Yeah, I he sort of went into his thinking, like, okay, I'm probably not going to solve this thing, but hey, let's just let's give it a shot.
• 4:45
  SPEAKER_03: I mean, I had all of the hubris of a teenager, right?
• 4:48
  SPEAKER_03: I wasn't even thinking that I was going to solve this thing.
• 4:51
  SPEAKER_03: Um and and was just wrong, uh, which, you know, I think there's a lesson in humility there.
• 4:57
  SPEAKER_03: Um, but I like although it sounds crazy to have some entire kind of PhD on one equation, um, or even one term of one equation.
• 5:06
  SPEAKER_03: But I in my head it made sense.
• 5:07
  SPEAKER_03: My thing is often like, how do you break down really big gnarly problems?
• 5:10
  SPEAKER_03: And the big gnarly problem is that quantum mechanics and general relativity disagree.
• 5:14
  SPEAKER_03: And I was like, well, okay, I'm not gonna just solve that starting from that big picture.
• 5:18
  SPEAKER_03: But like when a star collapses to become a black hole, the entropy equation starts as like a quantum mechanics kind of thermal equation.
• 5:27
  SPEAKER_03: But then for a black hole, it's a general activity equation.
• 5:29
  SPEAKER_03: I was like, well, that's just like like I just need to add one term to make these consistent across a discontinuity, and that felt like a very bite-sized chunk relative to the bigger problem.
• 5:38
  SPEAKER_03: So it at the time it actually seemed much more tractable than it was in hindsight.
• 5:43
  SPEAKER_01: Yeah, so I don't know if there's even a lesson there because I I'm I'm thinking about this a lot in terms of, you know, my own personal life and sort of going out and doing uh like volunteer work and something that at this point in my life that I've realized is like, oh, there are immediate things that I can do.
• 6:03
  SPEAKER_01: I think maybe we even talked about this last time, you know, for for me, it's uh we've got a food pantry and I can go do that and that that now.
• 6:13
  SPEAKER_01: And and it strikes me looking at your career that that's what a lot of this has sort of come down to is like what are what are some immediate problems or or issues uh or challenges that uh that I can actually work towards solving.
• 6:29
  SPEAKER_03: Yeah, and like when does it make sense that I'm the person to do it?
• 6:32
  SPEAKER_03: Right.
• 6:33
  SPEAKER_03: Like particularly at Google X, we had a bunch of interesting projects, some of which were in my portfolio.
• 6:39
  SPEAKER_03: And some of them I was like, this is a problem X should definitely do, but like I'm not the person to lead that project.
• 6:44
  SPEAKER_03: I have none of the really specific expertise.
• 6:47
  SPEAKER_03: Whereas wearable robotics was something where I was like, oh, well, you need a little bit of understanding of robotics, but you also need understand a little bit of the machine learning that controls and a little bit of the wearables that kind of embody the product angle.
• 6:59
  SPEAKER_03: You know, and so it was like actually that kind of suited my mix of skills much better.
• 7:03
  SPEAKER_03: Um yeah, and then even as you say with volunteering, like I recently got passionate about kind of maternal healthcare, child mortality.
• 7:09
  SPEAKER_03: And the single biggest driver of those things is poverty.
• 7:12
  SPEAKER_03: I was like, I'm not gonna, I'm not gonna like change that needle on a national scale.
• 7:17
  SPEAKER_03: Sure.
• 7:18
  SPEAKER_03: But there was some low-hanging fruit.
• 7:19
  SPEAKER_03: I was like, oh, like actually this is a tech problem that I think she could push the needle on.
• 7:22
  SPEAKER_03: Um yeah, so I'm all about where I can have the most kind of marginal impact given who I am, as opposed to just, you know, throwing things into the ether that that feel very nebulous.
• 7:32
  SPEAKER_01: Yeah, so maybe this is maybe this is a a a good place to sort of contextualize things for people who who aren't familiar with your background, would uh if we kind of triangulate what your expertise is, because you know, you've you you studied a lot of things, um, and and your resume is like quite frankly all over the place, right?
• 7:57
  SPEAKER_01: I mean, you'll ignore you'll acknowledge that.
• 7:59
  SPEAKER_01: Um so so so when you look at these challenges and and and when you look at ones that are you're really suited to take on, like what boxes do they tick?
• 8:11
  SPEAKER_03: Um I think it's often like I would say what I'm good at is first principles problem solving.
• 8:17
  SPEAKER_03: And that's kind of what physics gives you, right?
• 8:20
  SPEAKER_03: Um and I love learning new things, trying new things.
• 8:23
  SPEAKER_03: And so I'm attracted to challenges which are very interdisciplinary, because yeah, I'm never gonna be as good at motor control as our guy who spent the entire of his career on just that one thing.
• 8:34
  SPEAKER_03: But if your challenges are how do you balance in way different trade-offs across different disciplines, that's something that makes a lot of sense.
• 8:41
  SPEAKER_03: So if anything, I was attracted, yeah, studying, for example, I you say like at some point I did like a master's in development economics, which feels like way over here compared to all the physics work.
• 8:51
  SPEAKER_03: But I didn't know anything about that field at all.
• 8:55
  SPEAKER_03: And studying both gave me a lot of appreciation for like, okay, what are the kind of challenges that we're facing at a societal level or at a global level?
• 9:02
  SPEAKER_03: Um, and then which bits of that are kind of a good fit for some of the more technical background.
• 9:06
  SPEAKER_03: Um yeah.
• 9:08
  SPEAKER_03: I also think, you know, some of the stuff very opportunistic, right?
• 9:10
  SPEAKER_03: Like it's easy to now look back and paint kind of a narrative career.
• 9:14
  SPEAKER_03: I've also always met things that I find interesting in the moment, and then that kind of leads you on what looks like quite an unstructured journey.
• 9:22
  SPEAKER_01: Yeah, these are the sorts of things when you get to a point in your career when when you're feeling fairly successful that you're allowed to sort of like claim that they were somewhat structured or somewhat planned to get here.
• 9:33
  SPEAKER_03: Yeah, whereas like dropping out of the PhD at the time did not feel like a success, you know, even though in hindsight it was one of the best things I did.
• 9:41
  SPEAKER_01: Yeah, it it it's funny.
• 9:42
  SPEAKER_01: So uh Christina, who you just met, our our producer, uh, we brought her on fairly recently, and and she's new to this whole robotics world.
• 9:50
  SPEAKER_01: It's but something that I realized when I started covering it a while ago is is I think that a reason why I connected with it really deeply is something you're talking about, which is it it touches so many different things, um, you know, and and and it scratches so many different itches for me.
• 10:10
  SPEAKER_01: Like that I can I can have we we were in Boston two weeks ago, we did uh eight podcast interviews in four days, and every single conversation was dramatically different.
• 10:20
  SPEAKER_01: And that's something that's that's afforded us, and and it sounds like that's something that you've really been experiencing for as long as you've been in this world.
• 10:28
  SPEAKER_03: Definitely.
• 10:28
  SPEAKER_03: And also, I think very related to real problems in the physical world, which are very relatable to a broad audience as well.
• 10:36
  SPEAKER_03: Like one thing I love about the work we do, but I think it's true of a lot of robotics, is everyone gets it, right?
• 10:40
  SPEAKER_03: Like most robots are trying to solve a problem that is easy for someone to understand and and appreciate the value of, as opposed to more so than the entropy of the black holes.
• 10:50
  SPEAKER_01: Yeah.
• 10:51
  SPEAKER_01: Slightly more so.
• 10:52
  SPEAKER_01: I wanted to to uh talk about your like specifically your time at Google X, because prior to what ultimately became Skip, I mean, you were what what in uh in publishing we would call the slush pile, like you were in charge of going through like the the pitches, right?
• 11:10
  SPEAKER_01: And kind of like vetting some of these ideas that were coming through.
• 11:13
  SPEAKER_03: Yeah, exactly.
• 11:14
  SPEAKER_03: It was a job that's kind of part venture capital almost evaluating lots of ideas, part entrepreneur, you know, starting new projects and kind of pushing them to a point to make a decision.
• 11:24
  SPEAKER_03: Part like portfolio manager, you know, you have a bunch of baby projects and you decide which ones to invest in or not.
• 11:29
  SPEAKER_03: Uh yeah, it was a really interesting job.
• 11:31
  SPEAKER_03: Well, and what are you curious about?
• 11:33
  SPEAKER_01: I I I I well well, everything.
• 11:34
  SPEAKER_01: I'm I I guess I'm curious first like how you got that specific role and at the time or in hindsight, what you feel like in your background made you suited or qualified for that.
• 11:45
  SPEAKER_01: Because again, this is something that you can't be an expert in everything, but you've got to sort of know or at least be curious enough about things to go through and you know, either approve or disapprove of different pitches.
• 11:57
  SPEAKER_03: Yeah.
• 11:58
  SPEAKER_03: Um, I mean, almost exactly what we were just talking about.
• 12:01
  SPEAKER_03: So the the kind of rapid evaluation team, which is the one that I joined, um, historically has been super nerds, right?
• 12:08
  SPEAKER_03: So people who are serial PhDs, very deep tech, you know, very technical, excellent experts.
• 12:15
  SPEAKER_03: But I think part of the reason they brought me on board is that they were sometimes a little bit struggling with doing that triage or you know, part of part of what X tries to do is quickly kill projects.
• 12:26
  SPEAKER_03: And unfortunately, some people who are the kind of very technical narrow folk will get really excited about quantum levitation.
• 12:32
  SPEAKER_03: You know, it's like, oh my god, this new breakthrough have to do something about that.
• 12:36
  SPEAKER_03: And sometimes I think they were looking for someone who could go one level higher in the thinking and be like, okay, particularly at Google X, the mission was moonshots.
• 12:45
  SPEAKER_03: It's like you have to have a project that will move, like materially move alphabet valuation was kind of the goal.
• 12:52
  SPEAKER_03: And that's a high bar.
• 12:54
  SPEAKER_03: Like even like most startups where you can pitch a you know your billion dollar term, that's still not going to move alphabet valuation, right?
• 13:02
  SPEAKER_03: And so being able to quickly assess like, okay, does a new breakthrough in this particular area does it have the potential to be something that's like world-changing, requires a multidisciplinary approach, right?
• 13:15
  SPEAKER_03: Because we had a we had criteria, for example, for evaluating does the moonshop make sense?
• 13:19
  SPEAKER_03: And something for technical, right?
• 13:20
  SPEAKER_03: Like is the technical feasibility at the right stage?
• 13:23
  SPEAKER_03: And interestingly, part of that is often, is it too easy?
• 13:26
  SPEAKER_03: Because if something is too easy, it'll like Alphabet will lose to a quick startup who can hire faster, you know, has less regulatory burden, et cetera.
• 13:33
  SPEAKER_03: So you're looking for really hard problems but still solvable.
• 13:36
  SPEAKER_03: But then you also care about, okay, and are we creating enough value for the world that there's potential to capture value at scale, right?
• 13:43
  SPEAKER_03: Like we didn't have to solve the exact business model early on, but you had to solve a big enough problem to justify the investment.
• 13:49
  SPEAKER_03: Is it a problem that Alphabet is suited for?
• 13:52
  SPEAKER_03: Right.
• 13:52
  SPEAKER_03: So again, there are lots of other ideas which were interesting, but you're like, okay, does will it ever make sense for Alphabet to get into this space?
• 13:59
  SPEAKER_03: And there are some obvious no's, right?
• 14:01
  SPEAKER_03: Where it's like, if that's not a good fit for Alphabet, let's not gonna waste time on it now.
• 14:05
  SPEAKER_01: Um can you can you?
• 14:06
  SPEAKER_01: I mean, I I don't know how much you can like talk about the pitches, but can you give me any like examples of things that were like clearly not a fit for Google?
• 14:16
  SPEAKER_03: Yeah, I can give you examples of something that I worked on.
• 14:19
  SPEAKER_03: Um for one, I was passionate about uh external gestation.
• 14:23
  SPEAKER_03: So can you like incubate babies outside of the womb?
• 14:28
  SPEAKER_03: And primarily thinking about you know, either gay couples as an alternative to surrogacy, or yeah, or like, you know, there are many women who will get like a cancer diagnosis late stage in pregnancy, and then they have to pick between their own treatment or like aborting a baby that they really want, you know, like that's very stressful.
• 14:47
  SPEAKER_03: Um, and so the idea is, well, actually, current neonatal intensive care units are treating babies like small adults.
• 14:54
  SPEAKER_03: You know, they still have to have lungs that are capable of breathing, et cetera.
• 14:56
  SPEAKER_03: And you can instead replicate the uterus environment, you could do something really interesting there.
• 15:01
  SPEAKER_03: And so the tech was at the right level of difficulty, like it was hard, but kind of potentially kind of emergent.
• 15:07
  SPEAKER_03: But in hindsight, like, does Alphabet ever really want to open that kettle of fish?
• 15:11
  SPEAKER_03: Like, probably not, right?
• 15:12
  SPEAKER_03: Like it's not just kind of like ethically, you mean?
• 15:15
  SPEAKER_03: Oh, the range of reasons, right?
• 15:17
  SPEAKER_03: Like it would just be very far outside of Alphabet's portfolio.
• 15:20
  SPEAKER_03: And so that's all like a very big jump.
• 15:23
  SPEAKER_03: Um, another totally different example is I also was very passionate about informal markets.
• 15:28
  SPEAKER_03: So this is like particularly in lower income settings, a lot of small businesses or entrepreneurs, you know, that they're operating outside of a formal banking system, they're not regulated, their employment data is not captured, but partly as a result of that, they don't give access to finance and a lot of these things.
• 15:43
  SPEAKER_01: Well, yeah, and you and you I know that you prior to all this, like you you spent a good amount of time in in Africa, and a lot of uh a lot of the finances in Africa are those microtransactions that happen down there.
• 15:56
  SPEAKER_03: Exactly, exactly.
• 15:57
  SPEAKER_03: And in principle, we should be able to use kind of new data modalities to do a better job of understanding and providing services to businesses or kind of microentrepreneurs that don't currently get them.
• 16:09
  SPEAKER_03: Um and and that's kind of not necessarily a good fit for Alphabet in other ways, in the sense of okay, there's some tech there, there's some data, but fundamentally you're serving a population that is not like Alphabet's current core population.
• 16:21
  SPEAKER_03: There's a lot of nuance in the regulatory side.
• 16:23
  SPEAKER_03: Again, it just felt like it's a little bit further outside of Alphabet's wheelhouse than where they would like to sit.
• 16:29
  SPEAKER_01: I I'm just I'm thinking about the uh the external legislation thing and just like again, were Google to get into that world like man, like just just the the the press coverage, you know what I mean?
• 16:45
  SPEAKER_01: I mean, it would yeah, it would be all kinds of like just sort of like and and you know, and and I'm not casting any judgment at all, but you know, uh people's ideas of like what is and isn't dystopian when it comes to a large corporation getting involved in some of these sort of like forward-thinking, like you know, like borderline, like sci-fi ideas, like that's could be crossing a line for probably a lot of people.
• 17:09
  SPEAKER_03: Right.
• 17:09
  SPEAKER_03: And I kind of think that example is probably an obvious one that's a bad fit.
• 17:13
  SPEAKER_03: You know, I I can't talk about the ones that are maybe a little bit more borderline, but but yes, you can see that this is important.
• 17:19
  SPEAKER_03: Yeah.
• 17:19
  SPEAKER_01: Yeah.
• 17:20
  SPEAKER_01: So uh, I mean, obviously, ultimately you you hit upon um it's who is I I was talking to somebody the other week and they're like, Don't don't say wearables.
• 17:29
  SPEAKER_01: We don't say what we don't we don't say wearables anymore.
• 17:32
  SPEAKER_01: Or I don't I I don't know, I don't know if that's because like wearables are all like our consumer now, and the people who make like proper robotics don't like to refer to them as wearables.
• 17:41
  SPEAKER_01: But obviously, like you you landed in this kind of like soft exoskeleton or like wearable robotics world.
• 17:48
  SPEAKER_03: Yeah, we don't mind if you use the word wearables.
• 17:50
  SPEAKER_03: I mean we're trying to we're trying to push move wear, but at the end of the day, move wear.
• 17:54
  SPEAKER_03: We want to adopt the language that our customers understand, and fundamentally it is either powered clothing, powered pants.
• 18:00
  SPEAKER_03: Yeah, wearables is fine.
• 18:02
  SPEAKER_03: We're not pedantic.
• 18:03
  SPEAKER_01: Yeah, power again, powered pants might be problematic.
• 18:06
  SPEAKER_01: I don't know.
• 18:06
  SPEAKER_01: You know, we'll have to we'll have to workshop it.
• 18:10
  SPEAKER_01: Uh but uh I this my understanding is is for you again, this is something that came out of um some uh some personal experience.
• 18:20
  SPEAKER_03: Yeah, and so again, it's another reason why the team at X, I think, did a nice job of also sending diverse people, right?
• 18:26
  SPEAKER_03: So there were a lot of people that were very tech focused.
• 18:28
  SPEAKER_03: I was deliberately therefore trying to take a problem-focused approach to things.
• 18:32
  SPEAKER_03: So looking at what are the big, like global, huge problems, which will you are at the scale where you solve them, it's obviously a win.
• 18:39
  SPEAKER_03: Um and initially I was looking at aging.
• 18:42
  SPEAKER_03: And again, like older adult care, you know, is just underinvested in, right?
• 18:48
  SPEAKER_03: Like it's one of these spaces where the burden often falls on either women or like I mean older adults themselves, not often the kind of young Silicon Valley tech entrepreneur.
• 18:58
  SPEAKER_03: And so it just kind of hasn't had quite the attention that you might think is due, and especially given aging demographics, et cetera.
• 19:05
  SPEAKER_03: So I started a range of different projects which were all looking kind of at how do we make life more comfortable for older adults and in a scalable way, right?
• 19:14
  SPEAKER_03: So not this kind of you pay five million dollars for much of longevity treatments, but for kind of the average, median, 75, 85 year olds, how do we help them stay physically active and healthy as long as possible?
• 19:26
  SPEAKER_01: So you weren't thinking like how do we get teenage blood injections?
• 19:29
  SPEAKER_01: That wasn't necessarily the route you were going.
• 19:31
  SPEAKER_03: Yeah, we wasn't top of my list.
• 19:33
  SPEAKER_03: Okay.
• 19:34
  SPEAKER_03: Um, yeah, and now of course, Skip is a lot more general than just older adults.
• 19:38
  SPEAKER_03: So that was kind of where it started.
• 19:39
  SPEAKER_03: And thinking about, you know, yeah, like in my case, it was my grandmother, hit 90, wanted to live in her own home.
• 19:46
  SPEAKER_03: Um, she lived about a mile from the center of their little town.
• 19:49
  SPEAKER_03: And so she's she used to walk everywhere.
• 19:51
  SPEAKER_03: She'd walk to the coffee shop, she'd walk to a library, she'd walk to like flower club, you know, all the kind of cute things that little old ladies do.
• 19:57
  SPEAKER_03: Um one day she had a fall, lost the confidence to do that walk, and she just stopped doing it.
• 20:02
  SPEAKER_03: And I think the minute that that happened, there was a quick decline, both physically, right?
• 20:07
  SPEAKER_03: Because now she's not exercising and practicing, but also mentally, because like all of those social activities were kind of her, you know, her kind of mental well-being, her and her social community, and that started to get smaller as well.
• 20:21
  SPEAKER_03: And so I came to view physical movement and the ability to kind of be active and be a part of your community as essential for kind of physical and mental health care.
• 20:32
  SPEAKER_03: And there are just no good solutions, right?
• 20:34
  SPEAKER_03: Like, and that was very personal because I was kind of the tech one of the family trying to figure out okay, do we need to get granny or you know?
• 20:41
  SPEAKER_03: But the example of things on the market at the moment is like a button where after you've had a fall, you'll click the button and it'll text your loved ones.
• 20:48
  SPEAKER_01: I'm like life alert.
• 20:50
  SPEAKER_03: Yeah, and yeah, there's a few companies, but that's kind of the idea.
• 20:53
  SPEAKER_03: And that's great, it's better than nothing.
• 20:55
  SPEAKER_03: But what, like what low ambition, you know?
• 20:58
  SPEAKER_03: I don't want to get it, I don't want to get a text message after the fall.
• 21:01
  SPEAKER_03: I want to prevent the fall.
• 21:03
  SPEAKER_03: Like imagine if you had that approach to car crashes.
• 21:05
  SPEAKER_03: You're like, oh, we're not gonna bother with any safety features that prevent crashes.
• 21:08
  SPEAKER_03: We'll just like send you a text message after the crash has happened.
• 21:12
  SPEAKER_03: Um or similarly, there were lots of things where, okay, well, if you do a lot of rehab, that does improve your kind of long-term mobility, but it doesn't help you in the moment.
• 21:21
  SPEAKER_03: It's like if your goal is today, I want to get up, walk, you know, around my house, walk the dog, go visit a friend.
• 21:27
  SPEAKER_03: There was nothing that could actually help someone achieve those tasks.
• 21:30
  SPEAKER_01: And then the other side is like, you know, things that are like just replacing walking, right?
• 21:35
  SPEAKER_01: And then this wheelchair is outright, which are, you know, I mean, obviously they they they serve their function.
• 21:39
  SPEAKER_01: And obviously, there are people who just like probably will not be able to gain their mobility at all.
• 21:45
  SPEAKER_01: But it's not, you know, people who do walk a lot and and who have some potential to regain that mobility do want to walk again.
• 21:54
  SPEAKER_03: Right, right.
• 21:54
  SPEAKER_03: Or even you know, we talked to a lot of people who are maybe, for example, having Parkinson's as a kind of generative.
• 22:00
  SPEAKER_03: Degenerative condition.
• 22:01
  SPEAKER_03: And they often say their goal is to stay out of the wheelchair for as long, right?
• 22:06
  SPEAKER_03: Again, if you can keep physically active for even another three months, six months, that helps the rest of your kind of system uh stay in good shape.
• 22:14
  SPEAKER_03: And so even if there's a place for those products, like there's often a gap where you need a little bit of support, you need a little bit more than currently available, but you don't necessarily want to go to like a full wheelchair or all the time, right?
• 22:25
  SPEAKER_03: So again, maybe for some times you need the wheelchair, but the ability to have an alternative is also really powerful.
• 22:30
  SPEAKER_01: Yeah, I I really wanted to talk about the Parkinson's thing.
• 22:34
  SPEAKER_01: You and I have talked about this because again, this is something that's uh been personal in my life.
• 22:38
  SPEAKER_01: My my father was was diagnosed with Parkinson's last year or the year before.
• 22:43
  SPEAKER_01: It's something I've been looking into.
• 22:45
  SPEAKER_01: Um I think maybe the last time we spoke, we talked about I think I think it was a team at Harvard that was doing um, yeah, the piece on on the the Parkinson's research.
• 22:57
  SPEAKER_01: And something that I wasn't super aware of is is um a lot of it comes down to this like this freezing that happens, and it's sort of like a predictive gate that uh needs to to occur.
• 23:10
  SPEAKER_01: Um so like specifically when when looking at Parkinson's specifically versus other mobility issues, like you know, like MS, for example, uh what what are you what are you targeting?
• 23:23
  SPEAKER_03: Three different things.
• 23:25
  SPEAKER_03: So one is just general kind of strength and mobility, which will be very common across MS, across stroke, across other things.
• 23:32
  SPEAKER_02: Yeah.
• 23:33
  SPEAKER_03: And helping people with endurance, you know, particularly Parkinson's can be very tiring, right?
• 23:37
  SPEAKER_03: So just helping somebody walk a little bit further, a little bit more easily.
• 23:40
  SPEAKER_03: Bucket number two is freezing of gate.
• 23:42
  SPEAKER_03: So as you say, this is very specific to Parkinson.
• 23:44
  SPEAKER_03: And often it's it's only some people are at a particular stage in the disease progression where most people will be familiar with kind of tremors, where you could imagine if you have a tremor-like event in your legs, you kind of get stuck, right?
• 23:57
  SPEAKER_03: So you're trying to move forward, but you're mentally not sending the right signals from the brain.
• 24:00
  SPEAKER_03: And so you're you're literally frozen in place.
• 24:03
  SPEAKER_03: And for people who experience that on a daily basis, it can be debilitating.
• 24:08
  SPEAKER_03: Um, I mean, some of the kind of trials we do will get people to walk a circuit around their home.
• 24:13
  SPEAKER_03: And that might, without our assistance, might take them 10 minutes because every few steps, they'll kind of get stuck and almost like just waiting out.
• 24:21
  SPEAKER_03: Um and so that's something where you can really improve someone's quality of life by helping them kind of move through those moments.
• 24:28
  SPEAKER_01: And I and I assume that like in when you're in that kind of shape, like standing and waiting out is not a good option either.
• 24:36
  SPEAKER_03: Well, so this brings us to like bucket number three, which is balance and fall prevention, which I should say is the hardest and the one that you know I might get in trouble for talking about because we can't promise it will be able to prevent falls yet.
• 24:47
  SPEAKER_03: Um, but often when you are frozen, you know, one of our physical therapists will say that's when you can fall like a log, right?
• 24:53
  SPEAKER_03: Because if the freeze happens at the wrong moment, you can't kind of do a corrective step.
• 24:57
  SPEAKER_03: Yeah.
• 24:58
  SPEAKER_03: You're stuck in place and you see people kind of just all over.
• 25:01
  SPEAKER_03: And I'm sure everyone has, you know, someone in their family has had a fall.
• 25:04
  SPEAKER_03: Like these can be really painful events that can cause hospitalization from the start of a long decline.
• 25:09
  SPEAKER_03: And so, you know, that's kind of increasing levels of how confident we are, we can really make a difference.
• 25:13
  SPEAKER_03: Um, but ideally tackling all three.
• 25:15
  SPEAKER_03: And I think what's interesting and the part of the reason we started with Parkinson's is this freezing at gate is really salient, and we're actually really confident that we can improve that because we've seen amazing results, although currently on small numbers of people.
• 25:27
  SPEAKER_03: Um, whereas things like fall prevention is a little bit more speculative.
• 25:31
  SPEAKER_03: I think there's a lot more work we need to do on balance and falls.
• 25:33
  SPEAKER_03: But if we can show that for this population, there's also a good chance that we can show it for a much broader population.
• 25:39
  SPEAKER_03: And so someone with Parkinson's might wear our device for the freezing of gate, because that's a very daily constant experience for them.
• 25:45
  SPEAKER_03: But over time, we're kind of gaining training data, increasing our confidence, and improving balance, and over time making sure that we can do a better job at fall prevention as well.
• 25:54
  SPEAKER_01: Something that you were talking about when you you discussed the process of going through the pitches is that obviously, you know, at the end of the day, Google is looking for ideas that could be commercialized, right?
• 26:05
  SPEAKER_01: I mean, you know, that's what that's what a lot of this comes down to.
• 26:09
  SPEAKER_01: The reality, yeah, and and and you know, um, and you've had your experience in the VC world.
• 26:15
  SPEAKER_01: So like you know how you know how all of these things work.
• 26:17
  SPEAKER_01: I I'm curious, um, you know, when when the process really started to spin the company out uh and you were looking to commercialize things, how how are you determining what to actually uh productize and what segments of the markets to first go after?
• 26:35
  SPEAKER_03: Yeah, it's such a good question because a lot of the very early de-risking within X is focused on de-risking the like really big picture question, right?
• 26:44
  SPEAKER_03: Because it's not going to be this billion dollar you know market doesn't matter.
• 26:47
  SPEAKER_03: And so the early work within X, what answering questions like is there enough information in someone's gate to understand what movement they're doing before they do it?
• 26:56
  SPEAKER_03: Or you know, within kind of 10 milliseconds.
• 26:58
  SPEAKER_01: It's like predictive text for walking.
• 27:01
  SPEAKER_03: Exactly.
• 27:01
  SPEAKER_03: So we did a lot of these kind of language equivalents for the movement.
• 27:06
  SPEAKER_01: Yeah, well, like and and I mean, yeah, I mean, AI, right?
• 27:08
  SPEAKER_01: I mean, AI is probably we're pretty close to a place, if not there already, where we can probably predict that.
• 27:16
  SPEAKER_03: Yeah, so we have essentially foundation models of movement, which are structured in similar ways to what you'd have, you know, refination model for language.
• 27:23
  SPEAKER_03: Yeah, so we did a lot of that work at X, totally the void for many of the robotics, right?
• 27:27
  SPEAKER_03: Because you don't need a functioning exoskeleton to answer the question of is there enough information from worn sensors only to do this kind of prediction?
• 27:35
  SPEAKER_03: Um, and then would separately you know do a lot of user research around, okay, well, why do people not use walking sticks, right?
• 27:43
  SPEAKER_03: Like how how can we be sure that we're providing more value than that or easier to use, or are you what which sections of the market make sense?
• 27:50
  SPEAKER_03: And you know, why does a lot of work?
• 27:51
  SPEAKER_03: Why why don't people use walking sticks?
• 27:53
  SPEAKER_03: The biggest one is that it requires a hand.
• 27:56
  SPEAKER_03: And so if your goal is to go shopping, my granddad used to do this.
• 28:00
  SPEAKER_03: He would like hook his stick over his arm so he can carry something.
• 28:03
  SPEAKER_01: Yeah.
• 28:03
  SPEAKER_03: But obviously it's not very helpful.
• 28:05
  SPEAKER_03: The other big reason is stigma.
• 28:06
  SPEAKER_03: So again, this is something that we are very thoughtful around, is sometimes people and you know, I did this too, like you mentally don't want to admit that maybe you do need a little bit of support.
• 28:16
  SPEAKER_03: And so walking sticks and like walking frames require you to think I need a mobility aid.
• 28:24
  SPEAKER_03: And part of the reason our first product, MOGO, is actually more about hiking and aspiration, is we're trying to pitch it more like an e-bike.
• 28:30
  SPEAKER_03: And I think e-bikes are such a great analogy because they're for everyone and they can help anyone, and they're helping you achieve a task.
• 28:37
  SPEAKER_03: You're trying to go cycling, you're trying to climb this mountain.
• 28:40
  SPEAKER_03: But it happens that the people who probably get the most benefit out of e-bikes tend to be a little bit older, right?
• 28:46
  SPEAKER_03: So it's a good example where the marketing and the aspirational demographics are universal.
• 28:51
  SPEAKER_03: And so that means it's it's appealing and it's not stigmatized.
• 28:54
  SPEAKER_03: And yet the people who are really, really benefiting are often kind of perhaps a little bit further on that mobility spectrum.
• 29:00
  SPEAKER_01: It's sort of like, hey, you know, this is the the we're legitimizing this, just like this, this is something that anybody can use.
• 29:08
  SPEAKER_01: So it's not a sign of, yeah.
• 29:11
  SPEAKER_03: The other great examples of this are like the Apple earbuds, airpods.
• 29:15
  SPEAKER_03: Yeah, just the hearing aid.
• 29:16
  SPEAKER_03: They're a universal product.
• 29:17
  SPEAKER_03: Right.
• 29:17
  SPEAKER_03: But now they're also a hearing aid, so they're not as stigmatized as someone worrying about the hearing.
• 29:21
  SPEAKER_01: Yeah, it it is interesting.
• 29:23
  SPEAKER_01: And and you know, one of one of the one of the companies that we were talking to the other week um in Boston is uh doing a uh I I can't remember the name of the company, but they're they're they they are doing a kind of a modified wheelchair that's helping people get in and out of of the bed.
• 29:39
  SPEAKER_01: And you know, one of the one of the words they used was was dignity, right?
• 29:44
  SPEAKER_01: And and this idea of like you know, finding a lot of people want to continue to live independently.
• 29:51
  SPEAKER_01: And there's this idea of like of like aging in place.
• 29:54
  SPEAKER_01: And and and I I think I think what you're getting at here is not necessarily something where because I there was a company called Richter, I think.
• 30:05
  SPEAKER_01: I don't was it Richter?
• 30:06
  SPEAKER_01: I don't know if they're around anymore, but it was like a soft like underwear exoskeleton.
• 30:11
  SPEAKER_01: Um and I the the idea behind that was to stay like completely hidden.
• 30:17
  SPEAKER_01: Yeah.
• 30:17
  SPEAKER_01: But this your product isn't necessarily staying hidden because you're, you know, it's a a piece of exercise wear.
• 30:26
  SPEAKER_03: Right.
• 30:26
  SPEAKER_03: Like if anything you wanted to be aspirational and you know, my dad raves about his e-bike.
• 30:30
  SPEAKER_03: I want people to be raving about, oh, I tried the mogo, it really helped me, like, you know, telling their friends about it.
• 30:36
  SPEAKER_03: And it's it's also hard to hide something, right?
• 30:38
  SPEAKER_03: Like, of course we looked at when you do it for the underclothing, but then you've got some awkward bulges.
• 30:42
  SPEAKER_03: You know, that's that's I think worse because it's almost like admitting that your product's unattractive.
• 30:46
  SPEAKER_03: And instead we're trying to build a product that feels to people more like an e-bike or you know, an Apple product than one that feels like a mobility frame or a walking stick.
• 30:56
  SPEAKER_01: Yeah.
• 30:57
  SPEAKER_01: So okay, so so you were you were going through, you found that people didn't want to use walking sticks.
• 31:01
  SPEAKER_03: Yeah.
• 31:02
  SPEAKER_03: So we so in general, edX means a lot of this de-risking the big picture before we did any building of the kind of product.
• 31:08
  SPEAKER_03: But then at some point you say, okay, great, we're confident now it's a huge market, the tech's ready, it's in the right fit.
• 31:13
  SPEAKER_03: Like, what do we act?
• 31:14
  SPEAKER_03: Like, how do we start going in?
• 31:16
  SPEAKER_03: And again, even then we did a lot of looking at, well, a lot of companies go to logistics first, for example, and we didn't.
• 31:24
  SPEAKER_03: So two reasons.
• 31:24
  SPEAKER_03: One, I think long-term these jobs will be fully automated.
• 31:27
  SPEAKER_03: So it's actually like a shrinking market rather than a growing market.
• 31:29
  SPEAKER_03: Um, but two, it also doesn't give us the key learnings that we need in order to make the jump to the much bigger market, right?
• 31:36
  SPEAKER_03: Like if the big market is every man and their dog in the kind of aging population, like one in four people have some mobility challenge, and that's much bigger than the number of logistics workers.
• 31:45
  SPEAKER_03: But the hard part about that, retreating for that, one is it controls, because you've got to deal with people moving in their home or about their community, unstructured environments, very different from a logistics warehouse, which is level ground, indoors, controlled setting.
• 31:58
  SPEAKER_03: Um, and then the other thing is you have to solve all these human factors questions, right?
• 32:01
  SPEAKER_03: Like, how do we make it aspirational?
• 32:02
  SPEAKER_03: How do we make it a desirable product?
• 32:04
  SPEAKER_03: How do you ship it to someone and have them be able to put it on and adjust it within a minute or two?
• 32:08
  SPEAKER_03: And none of those things would we be proving and learning and kind of de-risking if we started in an easier kind of B2B setting, right?
• 32:17
  SPEAKER_03: So we then kind of narrow down and say, we want the beachhead to be consumer, aspirational, but also fairly robust population.
• 32:25
  SPEAKER_03: So we're not deliberately, you know, deliberately we're not starting with the kind of fragile older adults, we're starting with hiking and then over time kind of build out a more general pathway.
• 32:34
  SPEAKER_03: And so then, like at that point, we basically can say, great, now we can start building the first product.
• 32:38
  SPEAKER_03: And we did start some of that internally within X.
• 32:41
  SPEAKER_03: But I'll also say X was an incredible accelerant for our first year or two, right?
• 32:45
  SPEAKER_03: Because our ability to attract the top talent from like our first team was we asked every top US lab to send their best graduating PhD student.
• 32:54
  SPEAKER_03: So we had like an amazing team of roboticists, and then we pulled in a few folks from like DeepMind and like bigger Google who brought that to the table.
• 33:01
  SPEAKER_03: And you could you couldn't assemble such an interdisciplinary team on the outside, I think, as quickly and as easily, right?
• 33:07
  SPEAKER_03: Because it's such a complex problem.
• 33:08
  SPEAKER_03: You do need experts in all of these things.
• 33:11
  SPEAKER_03: Um so it was an amazing kind of getting us going and prototyping the risky.
• 33:16
  SPEAKER_03: There comes a point where, like, okay, Alphabet is maybe a slower, slightly more bureaucratic place than a startup.
• 33:21
  SPEAKER_03: So there's a point where maybe it slips from being an accelerant to, okay, we'd love to move a little bit faster, but you know, we have to be a bit more cautious.
• 33:28
  SPEAKER_03: And so we'd anyway been thinking, are we at the point where this might be, you know, quicker to do as a spin-out?
• 33:35
  SPEAKER_03: And also perhaps more palatable to Alphabet, right?
• 33:38
  SPEAKER_03: Like um, they were they've gone through phases about how much they want to be in the hardware game.
• 33:43
  SPEAKER_03: And I think at the same time, this was this is in was it 2022?
• 33:47
  SPEAKER_03: Yeah, I think it's on the public record that Alphabet was also looking at places where they think, oh, maybe this is we're less confident this would be a good fit.
• 33:54
  SPEAKER_03: Maybe this is a team that you know we can spin out and get kind of off the back of the state.
• 33:58
  SPEAKER_01: If you if you look at, if you look at uh and and again, I know you were just on Astroteller's podcast, and I think he's been he's definitely discussed this.
• 34:06
  SPEAKER_01: Like if you look at the X teams that like over the last two years that have spun out, I mean it's a tremendous frequency compared to you know decades before.
• 34:16
  SPEAKER_03: Yeah, and I think you know part of that has been alpha getting clearer around okay, what do they want to do themselves?
• 34:21
  SPEAKER_03: Or is this the best of both worlds where you know we can spin out and we can fall on our face if it doesn't work and it's you know, yeah, they have some equity in that world and let them do their thing.
• 34:31
  SPEAKER_01: Yeah.
• 34:32
  SPEAKER_03: Yeah, yeah.
• 34:33
  SPEAKER_03: Um yeah.
• 34:33
  SPEAKER_03: So I think it was it was a good confluence of timing where we were kind of itching to get outside of maybe move a bit faster, and then Alphabet was looking for teams that kind of they could spin out and move off books.
• 34:45
  SPEAKER_03: So yeah, that was the timing was convenient, let's say.
• 34:49
  SPEAKER_01: So you mentioned something uh that that's really interesting to me because I I we haven't talked about this and and it hadn't really occurred to me.
• 34:58
  SPEAKER_01: You know, you mentioned not going after logistics, not going after warehouses.
• 35:02
  SPEAKER_01: And if you look at a lot of really most of the big exoskeleton companies, there there is that break, right?
• 35:09
  SPEAKER_01: They've got the two sides of the company, they've they've got the the mobility, and then they've got the you know, for lifting boxes or for like doing construction and things, things like that.
• 35:18
  SPEAKER_01: But then you mention unstructured environments.
• 35:21
  SPEAKER_01: And unstructured environments, we have this conversation a lot around um home robots versus warehouse robotics.
• 35:27
  SPEAKER_01: That a warehouse is a structured or semi-structured environment, therefore, you know, navigation is easier than in a home because there's a lot of um nobody's no two homes are the same, you know, the lighting's all over the place, people leave their laundry, things like that.
• 35:43
  SPEAKER_01: Um but it didn't occur to me that that same thing might be playing out on an exoskeleton because obviously, like, you know, it's not it's not navigating the same way.
• 35:54
  SPEAKER_01: It's not, it doesn't actually have the contact with the floor, it's you know, not like rolling over the stuff.
• 35:59
  SPEAKER_01: So, how does whether or not it's uh a structured environment actually have an impact over something like an exoskeleton?
• 36:07
  SPEAKER_03: So traditional exoskeleton controls would be based on a few state machines.
• 36:12
  SPEAKER_03: So they would fine-tune a controller in the lab, probably on a treadmill, so it could do a nice job with walking, right?
• 36:18
  SPEAKER_03: Like it understands how much tool to apply over each step, and it feels really good for walking on a treadmill.
• 36:24
  SPEAKER_03: And then at best case, maybe you do the same for squatting or standing up out of a chair.
• 36:28
  SPEAKER_03: You know, so kind of the cutting edge, you know, outside of skip typically is they'll have a controller that's tuned for a few activities.
• 36:36
  SPEAKER_03: And as companies get bigger and better, they might add more activity.
• 36:39
  SPEAKER_03: So now they can do cycling as well.
• 36:41
  SPEAKER_03: But fundamentally, you're switching between activities.
• 36:43
  SPEAKER_03: Now, that's probably fine for a warehouse because realistically someone is either standing or they're walking in a straight line.
• 36:48
  SPEAKER_03: Like that's fine.
• 36:49
  SPEAKER_03: But as you start to think about, especially an older adult, people doing more interesting things for their body, like you know, just moving in different ways, leaning forward, kind of crouching in unusual circumstances.
• 36:59
  SPEAKER_03: You you can't do that.
• 37:00
  SPEAKER_03: Like you can't just have these set um kind of states or set activities.
• 37:05
  SPEAKER_03: And you know, a very saline example of this is some of our earlier controllers.
• 37:09
  SPEAKER_03: Uh, if you'd bend down to lift a heavy box, you want assistance doing that.
• 37:13
  SPEAKER_03: So it would turn on, it would help you.
• 37:14
  SPEAKER_03: But if I would try and bend down to tie my shoelace, it's very hard for some of these state machines to know the difference between tying a shoelace and that I'm about to pick up a heavy box.
• 37:22
  SPEAKER_03: So it tries to thrust you upright.
• 37:24
  SPEAKER_03: It's cool, I'm still trying to tie a shoelace.
• 37:26
  SPEAKER_03: Terrible user experience, right?
• 37:27
  SPEAKER_03: So I mean it really is in that control.
• 37:30
  SPEAKER_03: So we've had to go away from these kind of state machines into a more end-to-end machine learning approach, which requires a lot more data and a lot of unstructured inputs, but gets you something that will much more accurately, even in movements it's never seen before, understand what the human body is doing and therefore like what is the right type of assistance to apply.
• 37:49
  SPEAKER_03: That's the biggest one.
• 37:51
  SPEAKER_03: I'll also say uh the user needs and requirement change as you start serving high-need population.
• 37:57
  SPEAKER_03: So in a warehouse setting, maybe this vexes build and does only do listing or walking in a straight line.
• 38:03
  SPEAKER_03: And then if you do something unusual, it will just turn off.
• 38:05
  SPEAKER_03: And that's totally fine if you're helping an able-bodied worker be 5% more efficient.
• 38:10
  SPEAKER_03: But if you start to work with populations that are relying on your product, you can't just opt out of like 10% of the movements that they're doing.
• 38:17
  SPEAKER_03: The user is often expecting support all of the time.
• 38:20
  SPEAKER_03: And so the bar becomes much, much higher for okay, we actually do have to understand and provide a reasonable amount of support in every possible type of movement someone could do.
• 38:30
  SPEAKER_03: You can't just kind of turn off because you're not sure.
• 38:33
  SPEAKER_01: And then even like higher, if potentially you do get to a point when you you are able to support uh fall prevention.
• 38:40
  SPEAKER_03: Right.
• 38:40
  SPEAKER_03: Yeah.
• 38:41
  SPEAKER_03: Like you can't just turn that off.
• 38:44
  SPEAKER_01: Yeah.
• 38:45
  SPEAKER_01: Right.
• 38:45
  SPEAKER_01: Yeah.
• 38:46
  SPEAKER_01: Uh so you're you are you're building your own foundation models of human movement.
• 38:53
  SPEAKER_03: Yeah.
• 38:54
  SPEAKER_03: Yeah.
• 38:55
  SPEAKER_03: And so we we also probably have the world's largest data set of human movement.
• 38:59
  SPEAKER_03: And people will say, well, you can use video data and things like that.
• 39:02
  SPEAKER_03: But even the best kind of pose estimation models and video data will be plus or minus 10% on joint cinematics.
• 39:09
  SPEAKER_03: And that again, that's fine if you're just trying to guess like, is this person doing a star jump successfully?
• 39:14
  SPEAKER_03: But if you're trying to apply sources to the body, you have to be a lot more accurate than that.
• 39:18
  SPEAKER_03: And so, yeah, we have, I think, the world's largest data set is kind of on-body actuated information of how people, not just how people are moving, but then how they respond to actuation.
• 39:27
  SPEAKER_03: And that in it requires understanding of what's happening in the human biology.
• 39:31
  SPEAKER_03: So we're estimating kind of an internal like biological joint torque a lot of the time, as well as then what the appropriate assistance to a client is.
• 39:39
  SPEAKER_03: And it's got to work.
• 39:40
  SPEAKER_03: Yeah, so not just every environment that's spoken about, but also a huge variety of people, right?
• 39:45
  SPEAKER_03: Everyone walks a slightly different way, especially as you might get into serving populations that have a bit of knee pain or a bit of kind of a pathological gait, everyone walks with it differently.
• 39:54
  SPEAKER_03: And so the ability to understand that and also learn and adapt to different individuals and different settings becomes really interesting.
• 40:01
  SPEAKER_01: This is fascinating because, you know, like all I've been talking to people about over the last few months is um is is the data flywheel with robotics, right?
• 40:11
  SPEAKER_01: And and you know, the importance of collecting real-world data and real world versus video versus synthetic data and and teleop and and all these things.
• 40:21
  SPEAKER_01: And it sounds like maybe by necessity that you are and have to collect real-world movement data to train your models.
• 40:30
  SPEAKER_03: Yes.
• 40:31
  SPEAKER_03: And again, going back to why did we pick hiking, there's a lot of other reasons, but something that was on our mind, if not something we talk about as publicly, is that hiking is a lot of variability, right?
• 40:40
  SPEAKER_03: So if our goal is to collect as much real-world data, but also lots of different steps, sizes and shapes, you know, hiking is a very efficient way to get a huge variety um of movement data in the model, which as you say is what we need.
• 40:52
  SPEAKER_01: So to be clear, um, so right now, I and and you started this a while ago.
• 40:57
  SPEAKER_01: I still haven't had a chance to try it, but you are you're doing these rentals of the mogo where people can can can hike in them.
• 41:05
  SPEAKER_01: So but part of that experience is that you're collecting the data to train your models.
• 41:11
  SPEAKER_03: Yes.
• 41:12
  SPEAKER_03: And we run two different types of models.
• 41:14
  SPEAKER_03: And so understanding what the human is doing and understanding what's happening in the human biology.
• 41:19
  SPEAKER_03: What's great about that data is it's self-labeling.
• 41:22
  SPEAKER_03: As in what we're trying to do in the moment is a prediction task.
• 41:25
  SPEAKER_03: We're saying, okay, based on the first tiny fraction of a step, do we think you're going to go up a hill or down a hill or down some steps?
• 41:32
  SPEAKER_03: And we're trying to forward predict back in the map.
• 41:34
  SPEAKER_03: But after the movement has happened, we we have the labels from the sensors, right?
• 41:38
  SPEAKER_03: So we afterwards we kind of know what.
• 41:40
  SPEAKER_03: And so that is a great data flywheel because it's you know, unlabeled or self-labeling data that is making our you know uh models of human movement more and more accurate over time.
• 41:51
  SPEAKER_03: Yeah, so that's one part of it.
• 41:52
  SPEAKER_03: The other part is kind of understanding how different humans prefer assistance.
• 41:58
  SPEAKER_03: And that's a little bit more taxing to collect because we often will need to ask people like, oh, play around with a couple of these sliders or do some A-B testing to really dial in that.
• 42:07
  SPEAKER_03: And depending on whether you're booking one of our like paid experiences or whether you're signing up to a bit more rigorous data collection, uh, we'll do a mix of either just you're consenting to share your kind of unlabeled, you know, I and U data.
• 42:18
  SPEAKER_03: Uh, and most people don't really mind that, or we might do a bit more of an intensive data collection to also get some more kind of qualitative feedback on the types of assistance we're applying in those moments.
• 42:28
  SPEAKER_01: How did you get it to a point though, w where you were comfortable um doing this, doing these this testing or doing this trial like outside of the select group or outside of the company?
• 42:39
  SPEAKER_03: I mean, a lot of testing, right?
• 42:41
  SPEAKER_03: Like we've been at this for five years.
• 42:43
  SPEAKER_03: So you know, everything goes through like dyno testing, kind of on test rigs before it goes on to our team, and then our team will wear it out in the field for a lot before it will ever go on to someone who is a customer.
• 42:54
  SPEAKER_03: And then I would say we also have some um, you know, kind of a beta testing program where we also know that these people are fairly robust, right?
• 43:02
  SPEAKER_03: And so we'll kind of test it with them first and get a lot of feedback um before we release it to any member of the public.
• 43:07
  SPEAKER_03: And even then at the moment, we do a screening poll with people who bought an experience and just double check, especially double check that someone's not got too impaired or mobility, right?
• 43:16
  SPEAKER_03: So, like, this is a consumer product, this is a recreation device, we're not making medical claims for this first product.
• 43:21
  SPEAKER_03: And so part of the sad reality is sometimes people will apply because they're desperate for a solution.
• 43:25
  SPEAKER_03: There is nothing out there for cerebral palsy or for MS.
• 43:28
  SPEAKER_03: But at the moment we do unfortunately turn down someone, if then either their joint kinematics are just too different and we're not comfortable doing that yet, um, or if they need more support than we're confident our um controls can provide.
• 43:40
  SPEAKER_01: You don't want your product to do a disservice to somebody who's in, yeah.
• 43:43
  SPEAKER_03: And I mean sometimes in middle grounds, like sometimes we'll say, Hey, we don't think you're a good fit for our like multi hour hike.
• 43:49
  SPEAKER_03: If you want to come in and do a very short test where we have, you know, a PT or someone supervised, you know, there are sort of in between things we can do.
• 43:56
  SPEAKER_03: But yes, we're very thoughtful in what features we roll out and making sure.
• 44:00
  SPEAKER_03: That we're getting the right customers into these early uh demos.
• 44:04
  SPEAKER_01: So, how does clinical testing fit into all this?
• 44:07
  SPEAKER_03: So we do two chaps of clinical testing.
• 44:09
  SPEAKER_03: So one is with the Parkinson's work.
• 44:11
  SPEAKER_03: This is um done in support with Michael uh J.
• 44:14
  SPEAKER_03: Fox Foundation.
• 44:15
  SPEAKER_03: Um, and this kind of it's a you know, this clinical trials in IRB.
• 44:19
  SPEAKER_03: We do a lot of work with UCSS and with Emory and Georgia Tech.
• 44:22
  SPEAKER_03: You know, so again, there's often a physical therapist kind of running these sessions or will be done in a gate lab with a full harness.
• 44:28
  SPEAKER_03: Um so that's kind of a very specific, kind of slightly separate line of work.
• 44:32
  SPEAKER_03: And the Parkinson's device also has hit actuation, which is not something we have on Mogo.
• 44:36
  SPEAKER_03: So again, it's a different prototype and kind of a bit of a separate track.
• 44:39
  SPEAKER_03: Um, but then the other thing that we're just starting to do is the MOGO is a recreation device designed for hiking.
• 44:45
  SPEAKER_03: We've had so many people with knee pain say it's been really helpful that we're already starting on a next generation of that product, which we would hope to get FDA approval for to at least be able to make claims around knee pain.
• 44:56
  SPEAKER_03: This is a work in progress, we can't make those claims yet.
• 44:59
  SPEAKER_03: Um, but we're starting to do some early, you know, scoping out what trials there would look like.
• 45:04
  SPEAKER_03: And that's both kind of short term, we need to show that does reduce pain in our population.
• 45:08
  SPEAKER_03: But we also want to make sure we're tracking people over time so we you know can at least say, hey, use this for six months, twelve months.
• 45:15
  SPEAKER_03: It's going to improve your cartilage, just giving your chat your knee a chance to regrow.
• 45:18
  SPEAKER_03: And at least it's not causing any adverse events or any detriment down the line.
• 45:22
  SPEAKER_01: Yeah.
• 45:22
  SPEAKER_01: And obviously, rightfully so, you know, people's medical data is very uh, but there are a lot of protections around it.
• 45:30
  SPEAKER_01: Um but are you still able to use uh the the clinical testing to train models?
• 45:37
  SPEAKER_03: Yes.
• 45:37
  SPEAKER_03: So often de-identified data, right?
• 45:39
  SPEAKER_03: So none of this is a different thing.
• 45:40
  SPEAKER_01: Anonymous, yeah.
• 45:41
  SPEAKER_03: Yeah.
• 45:42
  SPEAKER_03: Um and then at the moment, uh, gate data is not necessarily seen as medical data.
• 45:46
  SPEAKER_03: So it's like if it's the device telemetry, that's more okay.
• 45:50
  SPEAKER_03: If we want to start you know saying, oh, for knee pain or for someone's Parkinson's, you know, then it's covered by like you know, strong regulations around health data, everything's anonymized, people have to opt into data sharing.
• 46:02
  SPEAKER_03: And what does FDA approval get you ultimately?
• 46:05
  SPEAKER_03: FDA approval gives us the ability to market that we do reduce knee pain.
• 46:09
  SPEAKER_03: That's what I mean.
• 46:10
  SPEAKER_03: Eventually we'll also be looking at insurance coverage, but you know, we're actually sold out of our first like nine months of manufacturing.
• 46:16
  SPEAKER_03: So, you know, at the moment it's less around getting the scale that would be unlocked with insurance coverage, and we're just allowing us to be really honest and targeted with people around here are the types of knee pain, we have evidence that it works for.
• 46:28
  SPEAKER_03: You know, here is but the hit the thing we don't as well, right?
• 46:31
  SPEAKER_03: So being able to be clearer around um where we think we can provide value and not.
• 46:36
  SPEAKER_03: Whereas at the moment it's just a consumer device and we're not making any medical claims right now.
• 46:40
  SPEAKER_01: Something I wanted to talk to you about, um, and you know, and this fits in with this whole uh drawing on um personal experience thing was something that came up the last time we spoke.
• 46:52
  SPEAKER_01: And please stop me if you're not comfortable talking about this.
• 46:55
  SPEAKER_01: I you know, I know you have discussed this uh publicly, um, but our last call was very interesting because it it had it happened like shortly after you had gone through a medical um uh issue.
• 47:11
  SPEAKER_01: But um ultimately um you know it it sounds like that has been uh a big driver uh for you to do this do this other work that you've also been doing in the meantime.
• 47:30
  SPEAKER_01: Uh Project Stardust is the name.
• 47:33
  SPEAKER_03: Yeah, yeah.
• 47:34
  SPEAKER_03: So to give listeners a bit of context, um I love my job at Skip, I'm founder CEO.
• 47:39
  SPEAKER_03: This is like 99% of my world.
• 47:41
  SPEAKER_03: And until a year ago, it was 100% of my world.
• 47:44
  SPEAKER_03: Um and then I was pregnant and you know, also like we did this hard fork podcast, which is one of our big press events, and I was nine months pregnant, but very visibly pregnant, right?
• 47:53
  SPEAKER_03: Like there aren't that many people who will Google Skip and not know that.
• 47:57
  SPEAKER_03: Um, and then our son died in childbirth at the end of a healthy pregnancy, no red flags, like no risk factors.
• 48:04
  SPEAKER_03: I was like, I I used to think of myself as more of an athlete, you know, the idea that death in childbirth is something that still happens standards.
• 48:11
  SPEAKER_01: One of the world's best hospital in, you know, in uh the US, like Yeah.
• 48:18
  SPEAKER_03: So just devastating.
• 48:19
  SPEAKER_03: Like, I mean, not on my radar at all, right?
• 48:22
  SPEAKER_03: And I, you know, ironically, I was one of these people, I was worried I wasn't gonna love a baby enough, you know.
• 48:27
  SPEAKER_03: Like, I'm not naturally a baby person, I'm kind of the like engineer.
• 48:30
  SPEAKER_03: Um, but of course, when you have a child, regardless of the outcome in this case, like you still have that intense feeling of like love and all of those hormones still come, and you're like, it I've never had such an intense, like that was both the best day of our life and obviously the worst, because he didn't make it, right?
• 48:49
  SPEAKER_03: Um anyway, so after some grieving period, initially I thought, okay, well, we'd put aside some money for childcare.
• 48:56
  SPEAKER_03: Could we do something to honor our son who's all ziggy?
• 48:59
  SPEAKER_03: Um, could we do something to kind of further you know the research or prevent this from happening to other parents?
• 49:05
  SPEAKER_03: And and started talking to experts.
• 49:07
  SPEAKER_03: And I'm so blessed, right, to be part of the standard community, part of the tech community here.
• 49:12
  SPEAKER_03: Our access to world-class experts was kind of unparalleled.
• 49:15
  SPEAKER_03: And so we spoke to a lot of people around who's doing good work, what does it take?
• 49:19
  SPEAKER_03: Like, and some of the things that we learned, in my mind, were shocking.
• 49:22
  SPEAKER_03: Like I'd be aware of something like SIVs, where you know you've got a baby lying on the back, and you know, you've got to watch out for them kind of being smothered.
• 49:31
  SPEAKER_03: Um, but it turns out that stillbirth and death and childbirth is seven times more common than that.
• 49:36
  SPEAKER_03: So it's about 25,000 babies a year that we lose like this.
• 49:40
  SPEAKER_03: It's actually the leading cause of infant mortality.
• 49:42
  SPEAKER_01: I think we said one in 150, I was reading in the Yeah, one in 150.
• 49:47
  SPEAKER_03: Exactly.
• 49:47
  SPEAKER_03: Um and yet we don't really talk about it.
• 49:50
  SPEAKER_03: I mean, it's partly because it it is hard to talk about sometimes.
• 49:53
  SPEAKER_03: Yeah.
• 49:54
  SPEAKER_03: But it's so common and even more sadly, like so preventable.
• 49:59
  SPEAKER_03: So, you know, it's hard to estimate how much is preventable, but people say between 40 and 60 percent of these could be able to.
• 50:05
  SPEAKER_03: And ours is a good example of that.
• 50:06
  SPEAKER_03: Like our son had been totally healthy and we didn't a checkup, you know, a few days before birth, we were at term.
• 50:13
  SPEAKER_03: Um, and he was born really small.
• 50:15
  SPEAKER_03: Like he was born kind of third percentile.
• 50:17
  SPEAKER_03: And actually, people say, well, if we'd done any ultrasound scans or kind of gross scan in that last trimester, they would have picked up on that and likely would have recommended early delivery.
• 50:26
  SPEAKER_03: Because I mean, the good news for science here is that we're now really good at looking after premature babies.
• 50:30
  SPEAKER_03: So actually, one of the most common interventions is, okay, if something's going wrong inside and we're not quite sure what, deliver the baby and we'll he'll have a really good chance of surviving.
• 50:40
  SPEAKER_03: But the problem is at the moment, because of cost and hospital capacity, only some women, or particularly high-risk women, get the kind of gold standard of care where you get a lot of these extra scans and extra monitoring.
• 50:52
  SPEAKER_01: I I'm sure the American healthcare system is also probably not helping.
• 50:56
  SPEAKER_03: Doesn't help.
• 50:57
  SPEAKER_03: Doesn't help.
• 50:58
  SPEAKER_03: But I you know, I would say this is a fairly universal problem, right?
• 51:01
  SPEAKER_03: Of like a lot of women start as being low-risk pregnancies, and it particularly in that first pregnancy, they'll experience a stillbirth, and after that, your high risk the first time people don't know.
• 51:10
  SPEAKER_03: And so part of what Project Startup's doing, which you know, for me is a little bit of a passion project, weekend hobby.
• 51:17
  SPEAKER_03: We've got a great CEO now that's kind of driving the day to day.
• 51:20
  SPEAKER_03: Um, but we're trying to look at can we give that level of care in the home and then to all women for a fraction of the cost.
• 51:27
  SPEAKER_03: And technically, it's absolutely possible because we've done this for things like cardiac disease, right?
• 51:32
  SPEAKER_03: Like the same wearable technologies, whether it's wearable ultrasound or you know, kind of wearable heart rate monitoring in other spaces of medicine already exists.
• 51:41
  SPEAKER_03: It's just that no one had kind of pulled together the product to do this well to pregnancy monitoring.
• 51:46
  SPEAKER_03: And part of the reason we're motivated to start something is my husband happens to be a professor of machine learning at Stanford.
• 51:53
  SPEAKER_03: And the two hard parts of this problem are pulling together a product that is wearable, that mums would enjoy using, that kind of, again, this is a very multidisciplinary problem.
• 52:01
  SPEAKER_03: There's probably some soft goods involved, very similar to a lot of what we've just gone through at SKID.
• 52:05
  SPEAKER_03: The other hard half of this problem is you're gonna get a bunch of noisy data from the baby, right?
• 52:09
  SPEAKER_03: How do you decide what's clinically relevant?
• 52:11
  SPEAKER_03: Like you don't want to worry people by kind of sending too much information, too many red flags.
• 52:15
  SPEAKER_03: That's exactly my husband's research.
• 52:17
  SPEAKER_03: And so we had a little bit of, well, geez, if we wouldn't start this company, right?
• 52:22
  SPEAKER_03: Like the serial founder plus the ML person.
• 52:24
  SPEAKER_01: Yeah, literally nobody in on earth is probably more qualified at the moment.
• 52:29
  SPEAKER_03: Yeah, which is, you know, and it's also don't have that many female founders in hardware, particularly who start a company after having kids, right?
• 52:37
  SPEAKER_03: Like again, at a bit of an age skew of founders.
• 52:39
  SPEAKER_03: Yeah.
• 52:40
  SPEAKER_03: And so, yeah.
• 52:41
  SPEAKER_03: So we started a little company on the side.
• 52:43
  SPEAKER_03: But like I said, there's now excellent kind of day-to-day management of that.
• 52:46
  SPEAKER_03: And so I'm on the board and helping out uh here and there.
• 52:50
  SPEAKER_03: And and I'm just so passionate about it because there are so many low-hanging fruit.
• 52:54
  SPEAKER_03: Like there's a lot of long-term science also that we want to improve around pregnancy.
• 52:58
  SPEAKER_03: But even just getting like universal access to basic screening and giving doctors the kind of information that they need to make some of these decisions, um, I think would save thousands of lives a year.
• 53:09
  SPEAKER_01: Yeah.
• 53:10
  SPEAKER_01: Well, for first, first off, Zucky Stardust.
• 53:13
  SPEAKER_01: I need to point out the obvious, which is one of the things.
• 53:15
  SPEAKER_03: I'm glad you got the reference.
• 53:17
  SPEAKER_01: Um second, uh what what what specifically what is the piece of hardware that you're working on?
• 53:23
  SPEAKER_03: So two product lines.
• 53:25
  SPEAKER_03: The first one will be direct consumer and it will replace kit counting.
• 53:29
  SPEAKER_03: So at the moment, anyone who's maybe been pregnant, the doctors will say, Oh, you should spend an hour a day counting the baby's kicks.
• 53:36
  SPEAKER_03: And as a working mom, I did not do this in my first pregnancy, right?
• 53:40
  SPEAKER_03: Like that's an insane standard of care.
• 53:42
  SPEAKER_03: Um, and it's very subjective.
• 53:44
  SPEAKER_03: And that is a very easy thing to actually pick up with um I and Us, pain set of sensors that we just skip, um, and acoustic microphones, right?
• 53:51
  SPEAKER_03: And again, you need to a little bit of science and machine learning here to differentiate between is the sound coming from the baby kind of moving inside the womb, or is this we're clothing rustling?
• 54:00
  SPEAKER_03: Eminently solvable, right?
• 54:02
  SPEAKER_03: And so that's kind of that'll be a low-cost product that we can do for tens of dollars, get out there very quickly, and it's not a perfect solution, but we'll at least equip woman with the knowledge of, okay, there is actually reduction in fetal movement.
• 54:16
  SPEAKER_03: I'm not imagining it, or I wasn't just too busy at work.
• 54:18
  SPEAKER_03: Or conversely, you know, show like that's one of the few things that is predictive of stillbirth, and if you can catch earlier, and also empower parents when you go to the hospital.
• 54:26
  SPEAKER_03: You can ask all these questions like when was the last time you felt the baby move, or was it a strong movement?
• 54:31
  SPEAKER_03: That's really hard to know.
• 54:32
  SPEAKER_03: In hindsight, if you can pull up your phone app and say, here's the trend data, I think we'll see a lot more um appropriate kind of decision making very quickly.
• 54:40
  SPEAKER_03: There's a second product we're working on that's wearable ultrasound, and that gives you a lot more of the information clinicians are after around baby kind of growth, development, fluid levels.
• 54:49
  SPEAKER_03: That's that's the idea of really taking what's currently requiring an in-hospital visit and is only available to a small number of people and making it much more universal.
• 54:57
  SPEAKER_03: That will probably be a couple of years behind because FDA-pre products will need more clinical trials.
• 55:01
  SPEAKER_03: Yeah, so the idea is to have both in parallel.
• 55:03
  SPEAKER_03: Um, partly because there's a bit of a data fiwable effect where you actually use the ultrasound data to train the models for the consumer product, right?
• 55:10
  SPEAKER_03: Because how you tell the difference between a real physical movement and an imagined one is often using the ultrasound, which is kind of high resolution.
• 55:18
  SPEAKER_03: So we'll do clinical trials for both at once, but we'll have a consumer product to market hopefully within a year or two, and then uh a prescription or FTA approved product that will take a bit longer that include ultrasound.
• 55:28
  SPEAKER_01: So so we're at about time, but um, you know, I something I wanted to really ask you about this project, um, because because I I think there are a lot of lessons to be learned here.
• 55:39
  SPEAKER_01: And you know, again, in reference to Google, we were saying, yeah, obviously um they're weeding out the products that like there isn't some commercial value for them.
• 55:48
  SPEAKER_01: And of course they are, they're they're a for-profit company.
• 55:51
  SPEAKER_01: And again, you know, you have this this background in in VC and the startups that VCs are backing are the ones that they think that are ultimately gonna pay off.
• 56:01
  SPEAKER_01: But as you know, as somebody with with with this experience and and with these these resources, I you know, I like I love the idea of you know, I know I'm sure you have like zero free time, but you know, spending some of those resources to work on something that is like really important, but is probably I mean not on its face a huge a huge moneymaker, or you're not certainly you're not going in it to make a t a ton of money.
• 56:34
  SPEAKER_01: Um and I'm I'm just curious if there's any any any advice or or knowledge that you can impart on people who do have these resources and this knowledge and who would like to move the needle on something, actually like add something good into the world, but maybe don't know how to take those first few steps.
• 56:56
  SPEAKER_03: Well, I think the first thing to say is uh often it's too simplistic to look at where people are currently spending money.
• 57:02
  SPEAKER_03: Because it's true of both Skit and Project Stardust that these are populations who don't have good alternatives right now.
• 57:09
  SPEAKER_03: But if you start digging into the user research, you know, one of my best friends we just visited, she was worried about feeling the baby move less.
• 57:16
  SPEAKER_03: She had to take half a day of work unpaid, drive a long way to the hospital, wait for hours to basically get a checkup and be told everything was fine.
• 57:23
  SPEAKER_03: Like she implicitly paid hundreds of dollars for that experience, but it's not neatly counted in a market analysis in a report anywhere.
• 57:32
  SPEAKER_01: Yeah, I and I and I I appreciate that.
• 57:34
  SPEAKER_01: I I guess part, you know, there is a part of me, and this is the the idealistic and probably wrong part of me that doesn't want to entirely rely on capitalism to be the driver to necessarily do some good in the world.
• 57:46
  SPEAKER_03: Well, and I think the broader point is like if you're solving real problems for people, then often the way that you might capture value and prove that a business model can come later, right?
• 57:56
  SPEAKER_03: Like if you build a product that is solving a need that people love, yeah, you'll be creating value.
• 58:00
  SPEAKER_03: I mean, Waymo is such a good example of that.
• 58:02
  SPEAKER_03: They started Waymo before Uber existed, right?
• 58:04
  SPEAKER_03: So they didn't really have a good business model initially.
• 58:07
  SPEAKER_03: But by the time the tech was ready, there was an obvious one.
• 58:10
  SPEAKER_03: Um, and so yeah, I think the the more general advice point is if you're solving a huge global problem, you will be creating value.
• 58:17
  SPEAKER_03: And it will then be a question of how you capture that value.
• 58:20
  SPEAKER_03: But you can a little bit punt that, you know, or you can convince people, of course, there'll be a way to capture it because look at all this value we're unlocking.
• 58:26
  SPEAKER_03: And you just have to figure out who is benefiting from that and who and who has a little bit of money to chip in, right?
• 58:31
  SPEAKER_03: You know, another good example for both souls and for kind of childbirth is even if things go wrong, it's really expensive, right?
• 58:39
  SPEAKER_03: So actually the insurance companies often will pay tens of thousands of dollars for a lot of hospital stays, et cetera.
• 58:45
  SPEAKER_03: And it will take some time and some data to show that we can prevent souls or that we can prevent the complications of childbirth.
• 58:51
  SPEAKER_03: But if you can get that data, then you've unlocked a huge kind of value factor opportunity there as well.
• 58:57
  SPEAKER_03: Um but yeah, if you're solving real problems, that I think is much more important than the kind of initial market size uh that any DC will put on it.
• 59:05
  SPEAKER_01: Well, uh, Catherine, always a pleasure.
• 59:08
  SPEAKER_01: Thank you so much for taking the time.
• 59:10
  SPEAKER_03: Yeah, of course.
• 59:11
  SPEAKER_03: Thank you for having me.
• 59:13
  SPEAKER_01: So this is gonna be an addendum to the uh interview that we did with Catherine Zeeland.
• 59:18
  SPEAKER_01: Uh, she actually set up this conversation.
• 59:20
  SPEAKER_01: She's she's worked with you on trials in the past.
• 59:24
  SPEAKER_01: How did she first cross your path?
• 59:26
  SPEAKER_00: Yeah, so um I should intro and say I'm an assistant professor at UCSS.
• 59:31
  SPEAKER_00: Um, none of my views are representative of theirs.
• 59:34
  SPEAKER_00: Um, but basically, speaking for myself, I should say.
• 59:37
  SPEAKER_00: But essentially, uh, when I was doing my PhD under Dr.
• 59:40
  SPEAKER_00: Doris Wong, who's a neurosurgeon at UCSF, she um was partnered with Skip and Catherine's company to work on sort of their very early phase clinical trial.
• 59:51
  SPEAKER_00: Um, and so that was sort of how I became part of the project.
• 59:54
  SPEAKER_00: Then I'm a physical therapist by training before my PhD.
• 59:58
  SPEAKER_00: So was helping out doing a lot of selection tests and measures and protocol and guarding and all things.
• 1:00:03
  SPEAKER_00: And so um that was sort of my initial exposure.
• 1:00:06
  SPEAKER_00: And I think Skip at that time was probably an N of three or four um staff.
• 1:00:11
  SPEAKER_00: And so it's been really fun.
• 1:00:13
  SPEAKER_00: Um, you know, I I still work with them on and off and seeing how big they've gotten since then.
• 1:00:18
  SPEAKER_00: But that was our initial um exposure.
• 1:00:21
  SPEAKER_01: What what does working with them entail?
• 1:00:24
  SPEAKER_00: Um well, so in the past, I would say, you know, it kind of ebbs and flows depending on what phase they're at and funding and alignment of kind of other research projects I have going.
• 1:00:35
  SPEAKER_00: But um, you know, it it kind of basically right now it's been helping them with um data collection.
• 1:00:41
  SPEAKER_00: So they're working.
• 1:00:43
  SPEAKER_00: I I guess I should have probably said, I don't know how much Catherine said, so I don't want to totally maybe say too much, but basically, or or maybe you can speak to kind of some summary of what she gave of what they're currently working on.
• 1:00:55
  SPEAKER_01: Um yeah, I mean, I you know, this this this had come up, you know, we we've we've spoken about this a few times over the past year, um, especially since it, you know, it had become something uh personal to me for reasons we discussed earlier.
• 1:01:10
  SPEAKER_02: Yeah.
• 1:01:11
  SPEAKER_01: Um and and and I had I had done a little bit of um coverage around, I think there's a team at Harvard that's doing something similar as far as uh like gate prediction goes with Parkinson's.
• 1:01:22
  SPEAKER_01: So I I'm I'm I'm aware of that at very least, that that that it's something that they're kind of encompassing as well.
• 1:01:28
  SPEAKER_00: Got it.
• 1:01:29
  SPEAKER_00: Yeah, so um I guess just a kind of broad overview then.
• 1:01:32
  SPEAKER_00: So a lot of working with them is um getting helping them get the device on people, looking at how it's performing, um, you know, making sure people are safe, making sure the tests make sense that we're doing, you know, wanting to be as rigorous as possible with sort of experimental design and how we block, and um also make sure participants have like a good time too.
• 1:01:54
  SPEAKER_00: Is it mimicking the things they would see themselves doing with it?
• 1:01:58
  SPEAKER_00: Um, does it help?
• 1:01:59
  SPEAKER_00: Does it hinder?
• 1:02:00
  SPEAKER_00: Um, so it's been a nice kind of mix of what we call like mixed methods, so both qualitative and quantitative data analysis.
• 1:02:06
  SPEAKER_00: Um so that's depending on the phase, you know, it can be one-ops, it can be part of an overarching clinical trial by itself.
• 1:02:14
  SPEAKER_01: Do you find like generally speaking that the people, the um uh the the clients or I guess the patients that you're dealing with are generally excited to be trying out new things?
• 1:02:24
  SPEAKER_00: Yeah, I think it's really awesome because I think even it's fun for me as a physical therapist because I still practice clinically.
• 1:02:30
  SPEAKER_00: And I feel like, you know, people have often the sense of like, oh, I should do physical therapy.
• 1:02:36
  SPEAKER_00: I should do this thing.
• 1:02:37
  SPEAKER_00: You know, if Parkinson's often they've been told the things, right?
• 1:02:41
  SPEAKER_00: And so I think there's an amount of like buy-in that we sort of have to meet the patient with.
• 1:02:46
  SPEAKER_00: Whereas I think on the device side, it's kind of fun.
• 1:02:48
  SPEAKER_00: It almost always tell patients kind of like the wild, wild west a little bit where it's like industry side, it's almost like a playground, you know, come in, try some things on.
• 1:02:57
  SPEAKER_00: If something's not working, they can fix it on the fly and we'll go again.
• 1:03:00
  SPEAKER_00: And it's a lot more nimble, and I think that gets patients excited, both to get to contribute to research, but also I think it's a little more dynamic and they it's a little more novel and innovative um than like going to a PT clinic and doing all the things that they sort of expected that they would do.
• 1:03:16
  SPEAKER_01: Yeah, so so you know, I again I've covered Skip for a number of years now.
• 1:03:20
  SPEAKER_01: Um, and uh a lot of um a lot of what it does is sort of that that assistive walking, right?
• 1:03:25
  SPEAKER_01: So they're they're they're doing hiking.
• 1:03:27
  SPEAKER_01: Um so it makes sense from that standpoint.
• 1:03:29
  SPEAKER_01: But uh can you break down like what is specific to Parkinson's where a product like this might ultimately make sense?
• 1:03:38
  SPEAKER_00: Yeah, so I think there's a number of reasons.
• 1:03:41
  SPEAKER_00: I think number one is um, you know, when we talk about quark or cardinal symptoms of Parkinson's, we have like bradykinesia.
• 1:03:47
  SPEAKER_00: So that slowing of movement, um, force production can be difficult, timing, initiation.
• 1:03:52
  SPEAKER_00: So I think a device, you know, that basically augments somebody's sort of existing capabilities, right?
• 1:03:58
  SPEAKER_00: We're not talking about like paralysis where you have to basically do 100% of the work in every muscle group to support movement.
• 1:04:05
  SPEAKER_00: Rather, it's sort of this thought of like augmenting what they already have.
• 1:04:09
  SPEAKER_00: Maybe it's a timing thing and normalizing the timing out, maybe it's co-contraction of muscles around, you know, to stabilize, but you don't have to do 100% of the movement.
• 1:04:18
  SPEAKER_00: And I think that's an exciting engineering problem and also a trickier one.
• 1:04:23
  SPEAKER_00: Um, and so I would say that that's been um sort of, I think, the really big impetus for getting exos on folks with Parkinson's.
• 1:04:32
  SPEAKER_00: Um, and then there is a symptom called freezing of gate.
• 1:04:35
  SPEAKER_00: Basically, imagine like you have Parkinson's, you go into a grocery store around, you know, and I live in San Francisco.
• 1:04:41
  SPEAKER_00: So, you know, we have a lot of folks that'll go into their community grocery store, three houses down, they've been there forever.
• 1:04:48
  SPEAKER_00: They go in, there's tight narrow hallways and racks, dim lighting, lots of people and things, they're looking for items, all that stimulus, basically.
• 1:04:57
  SPEAKER_00: They'll be walking along and all of a sudden they just physically can't go anymore.
• 1:05:00
  SPEAKER_00: They're arrested in place, they can't take another step.
• 1:05:03
  SPEAKER_00: And where people get into trouble more is they start to lean forward, trying to use momentum to carry themselves to force that next step, but the next step doesn't come.
• 1:05:12
  SPEAKER_00: So now all of a sudden you have a really bad fall on your face, nose, things like that going down.
• 1:05:18
  SPEAKER_00: And so, as you can imagine, that's really humiliating and disturbing when you're in the grocery store and people are going by and they're not sure what's happening.
• 1:05:26
  SPEAKER_00: And dangerous.
• 1:05:27
  SPEAKER_00: So it has like all these elements to it that just make it really terrible and also it's really hard to treat it.
• 1:05:34
  SPEAKER_00: Um we don't totally know still what causes it on like a circuit level.
• 1:05:38
  SPEAKER_00: There's lots of different therapies, the brain stimulation, medication, PT.
• 1:05:42
  SPEAKER_00: No one quite has a secret sauce yet.
• 1:05:44
  SPEAKER_00: Um, so I do think from an EXO perspective or kind of a tech innovation perspective, there's a tremendous amount of opportunity to get some of the force production and timing and some of that to also work on freezing of gate, which is I think like one of the most exciting.
• 1:06:00
  SPEAKER_00: Exciting avenues for this line of work.
• 1:06:02
  SPEAKER_01: Yeah, I was reading an interview that you did um not super long ago, and um you were discussing as as as you and I uh talked about earlier, um you know, we were both interested in this for for uh personal reasons.
• 1:06:18
  SPEAKER_01: Um I guess what surprised me is how much how little is understood ultimately as far as how these the neurodegenerative diseases impact the circuits in your brain and how that affects the way the way you move.
• 1:06:33
  SPEAKER_01: I mean, that that's still an unsolved issue.
• 1:06:37
  SPEAKER_00: Yeah, yeah, pretty much.
• 1:06:39
  SPEAKER_00: I mean, I I have to give a lot of credit to there's a lot of really, really smart people, much smarter than me that are working on tackling this.
• 1:06:47
  SPEAKER_00: Um, Dr.
• 1:06:47
  SPEAKER_00: Wong, especially at UCSF, is using brain recordings pairing with like um biomechanics to try to get at this, you know.
• 1:06:55
  SPEAKER_00: So there's a lot of people that that realize that this is still a huge gap and are working on it.
• 1:07:00
  SPEAKER_00: But I mean, as of today, right now, there's still a tremendous amount we don't know.
• 1:07:03
  SPEAKER_00: I think if we knew more, it would be a lot easier to, whatever avenue is your choice, to kind of, you know, hey, this person has issues when they do this, we just need to do this thing to prevent it.
• 1:07:13
  SPEAKER_00: And we're just not quite there yet.
• 1:07:15
  SPEAKER_00: But I think it's also an exciting kind of opportunity and going back to your question earlier of our patients, you know, what is the patient side of things?
• 1:07:23
  SPEAKER_00: And I think that to them, patients that struggle with freezing a date, especially, they're like ready to do anything.
• 1:07:28
  SPEAKER_00: I mean, um, to to give an anecdote from a patient maybe a couple of weeks ago, they were saying how they will wait to cross the street, you know, the light timer comes on, and just that pressure that comes on when the light turns is so much that that causes them to freeze.
• 1:07:45
  SPEAKER_00: And sometimes it will take them eight to 10 to 12 light cycles before they can walk across the street, which like I think about that because I think that motivates all of us to work on freezing, whether you're a neurosurgeon, researcher, you know, as a PT, where we're like, this is such a huge issue, and like we have to figure something out.
• 1:08:03
  SPEAKER_00: And so I think um, yeah, it is crazy how little we know, but how many people are definitely working on it right now, whatever means they can.
• 1:08:10
  SPEAKER_01: Yeah, I this this is probably a question better asked of Dr.
• 1:08:14
  SPEAKER_01: Wong, but insofar as you're like comfortable talking about it, when you say brain recordings paired with biomechanics, what what does that look like?
• 1:08:24
  SPEAKER_00: Yeah, so basically um there is technology which exists that um when I was doing my PhD, we were working with a Medtronic device where they make specialized um brain implants where basically the DBS or the deep brain stimulation leads that are implanted in the brain, which sometimes people get for therapy for their symptoms.
• 1:08:45
  SPEAKER_00: Um they can actually record ambulatory neurophysiology recordings.
• 1:08:49
  SPEAKER_01: Yeah, I was I was looking into this before this conversation.
• 1:08:52
  SPEAKER_01: Um, and somebody, I don't know if you feel like this is an appropriate analogy, but somebody called it a pacemaker for the brain.
• 1:08:58
  SPEAKER_00: Yeah, totally.
• 1:09:00
  SPEAKER_01: It sends it sends like stimulation to the brain in order to directly in.
• 1:09:05
  SPEAKER_00: These um kind of more novel devices can record as well.
• 1:09:08
  SPEAKER_00: So deliver stimulation in and get signal coming out, you can record while somebody walks, runs, does all these tasks that people with Parkinson's often have issues with.
• 1:09:19
  SPEAKER_00: So by kind of correlating, you know, the behavior, okay, what does somebody's force production or timing look during their walking with their brain signals?
• 1:09:27
  SPEAKER_00: That's kind of trying to get it sort of how do these two relate as it goes to developing therapeutics and things.
• 1:09:33
  SPEAKER_01: Yeah, I was I was looking around some of the papers that um that you co-published.
• 1:09:38
  SPEAKER_01: And uh I anybody who has got I've I I've gone through PT.
• 1:09:42
  SPEAKER_01: I think most people Christina's definitely gone through PT, I think.
• 1:09:45
  SPEAKER_01: Um but uh folks who have gone through PT are familiar with the uh recumbent stepping machine.
• 1:09:53
  SPEAKER_01: Like it's like a recumbent bicycle, but you know, you actually do the steps.
• 1:09:57
  SPEAKER_01: Um and I I I'm curious, like that again, that's a staple of of PT, generally speaking.
• 1:10:04
  SPEAKER_01: Um what your role and your background as as a physical therapist, like what you're bringing to this conversation.
• 1:10:11
  SPEAKER_00: I think first and foremost, I go back to my PT training as, you know, at the end of the day, I think what excites me about PT, you know, obviously I was a lifelong athlete, I have this story about why I got a PT, but you know, I think it really came down to A, exercise, like at the end of the day, you know, if we're meeting minimum sort of principles of we need a stimulus, we need it to be progressive, things like that, right?
• 1:10:34
  SPEAKER_00: If that's done, I think what's cool is we don't sometimes need to worry so much about the nitty-gritty, like with the pharmaceutical, is it the right one?
• 1:10:41
  SPEAKER_00: Are there a lot of side effects?
• 1:10:43
  SPEAKER_00: Is this causing unintended issues?
• 1:10:45
  SPEAKER_00: Or with deep brain stimulation, we were talking about the pacemaker of the brain, sort of, you were giving that analogy.
• 1:10:50
  SPEAKER_00: It's invasive brain surgery.
• 1:10:52
  SPEAKER_00: And are the settings correct?
• 1:10:54
  SPEAKER_00: Do we have to reprogram?
• 1:10:55
  SPEAKER_00: How does that work with our medication?
• 1:10:56
  SPEAKER_00: I think with exercise, it's exciting to me because I'm like, I'm gonna give this to you.
• 1:11:01
  SPEAKER_00: And assuming I've done my due diligence to make sure you're safe and competent and your vitals are okay, you're gonna benefit.
• 1:11:07
  SPEAKER_00: And I think um that's really exciting to me, and then lets me sort of let the patient, you know, they start feeling better, the mood booster, you know, we get all these other benefits that start to bring them into the conversation.
• 1:11:17
  SPEAKER_00: And so I think that's really exciting to me in terms of what I bring to a conversation, is usually like, yes, there's an element of precision, but also I think it's fun for me to be like, I'm in this avenue where generally people are gonna get better when they come to see me or have a benefit, and I'm less worried about all these other pieces of things that might interact or might be invasive or so on and so forth.
• 1:11:40
  SPEAKER_00: Um, and there's great evidence for exercise in Parkinson's and a lot of disease modification pieces to research coming out now, like when people start this early enough, um, you know, we really do see a change trajectory of the disease.
• 1:11:53
  SPEAKER_00: So I think that's exciting.
• 1:11:55
  SPEAKER_00: I think the other element though, and I think with Skip especially, is because I was trained as a PT first, everything we do is sort of thinking about the patient.
• 1:12:03
  SPEAKER_00: So, what will improve quality of life?
• 1:12:05
  SPEAKER_00: What will improve their function?
• 1:12:07
  SPEAKER_00: Can they go to the grocery store now independently?
• 1:12:10
  SPEAKER_00: Can they get up from the floor if they were to fall?
• 1:12:12
  SPEAKER_00: Do they need um support?
• 1:12:14
  SPEAKER_00: What does that look like?
• 1:12:15
  SPEAKER_00: You know, that's those are all the questions that we ask.
• 1:12:17
  SPEAKER_00: And so I think coming to a device study, especially working with engineers and people, it's often, you know, I'm asking the questions of like, would they even use this?
• 1:12:26
  SPEAKER_00: Would this increase their function?
• 1:12:27
  SPEAKER_00: Do they care?
• 1:12:29
  SPEAKER_00: You know, like engineers can design a great device and then sort of approach the patient and they're like, I would never wear that.
• 1:12:35
  SPEAKER_00: It's too heavy, or it's ugly, or um, you know, it scrapes my arm every time I sit.
• 1:12:40
  SPEAKER_00: So I would never wear it out.
• 1:12:42
  SPEAKER_00: You know, like those questions I think sometimes they do get answered eventually, but I think having a PT, I'm I'm that's all I think about and observe and and ask.
• 1:12:50
  SPEAKER_00: And so I think um that's a lot of what I bring to their specific work.
• 1:12:55
  SPEAKER_01: Does being at UCSF like is are you kind of like uniquely positioned as far as like your proximity, obviously San Francisco, but also Silicon Valley, to um work with some of these tech companies?
• 1:13:06
  SPEAKER_00: Yeah, I think we get pitched a lot.
• 1:13:08
  SPEAKER_00: Um, in fact, I would say as a newer faculty member, newer researcher, um, I really learned to I think be strategic in sort of um in some of the science, especially.
• 1:13:20
  SPEAKER_00: I think Skip has done an outstanding job at doing all the right things, like hiring brilliant people, going through, you know, partnering with an academic institution, partnering with others, um, thinking about the patients you want to recruit, what a success looks like.
• 1:13:36
  SPEAKER_00: Um, you know, I think they've just done a lot of really, really amazing things.
• 1:13:40
  SPEAKER_00: I think some companies, they come to us and they're like, hey, you know, it's usually an engineer, hey, I made this product.
• 1:13:47
  SPEAKER_00: Can you try it in clinic and let me know what you think?
• 1:13:50
  SPEAKER_00: And, you know, they'll pitch it.
• 1:13:51
  SPEAKER_00: And it's sort of tricky for us because I think it's like a little bit of a mismatch in that like if I have a limited amount of time with a patient, do I have the bandwidth to try your product?
• 1:14:00
  SPEAKER_00: Like, what's the angle for me, especially if like I'm not being compensated or I'm not being trained on it, or you know, all these pieces.
• 1:14:07
  SPEAKER_00: And then I think it's tricky for them because like usually we'll use it for two days if it ends up in the workflow, and then it gets cast aside for it needed to be charged, nobody charged it or whatever.
• 1:14:17
  SPEAKER_00: And then all of a sudden, they don't, you know, we're like, oh yeah, we didn't use it.
• 1:14:21
  SPEAKER_00: And then that might be a win, you know, in terms of like it could have been a great device, but not the right use or not, we weren't the right setup for them.
• 1:14:28
  SPEAKER_00: And then it goes by the wayside too.
• 1:14:30
  SPEAKER_01: Yeah, that and that that's interesting.
• 1:14:31
  SPEAKER_01: So, so in terms of the trials that you're doing, it it you're really just finding a way to incorporate incorporate this into your kind of existing treatments and workflows.
• 1:14:41
  SPEAKER_00: It really depends.
• 1:14:43
  SPEAKER_00: I mean, so like with Skip, Skip did formal clinical trial, things like that.
• 1:14:46
  SPEAKER_00: So everything was outside of like a clinical workflow.
• 1:14:49
  SPEAKER_00: Um, but some 100%, I would say probably once every couple of weeks, we get pitched.
• 1:14:54
  SPEAKER_00: I get an email to my inbox, you know, saying, hey, I have this device.
• 1:14:59
  SPEAKER_00: Would you be willing to meet with me?
• 1:15:00
  SPEAKER_00: I'll show it to you.
• 1:15:01
  SPEAKER_00: And their goal then would just be to kind of have us trial in clinic.
• 1:15:05
  SPEAKER_00: Um, and so obviously, based off where it's at FDA regulation and our, you know, sometimes it never even makes it there, but it is sort of a whole workflow where sometimes that is their goal of just kind of getting some one-off data collections in a way that they didn't need to pay for consulting services or pay for an IRB or, you know, go through all of that, which is not my preferred route.
• 1:15:26
  SPEAKER_00: And I would say I think like going back to skip, they've done everything sort of from the start and this more work, but I think really high, high um benefit and really getting at what they wanted to get at versus just kind of meeting with us and trying out something.
• 1:15:41
  SPEAKER_01: How transformative have things like Fitbit and Apple Watches been as far as people's willingness to embrace some of these technologies?
• 1:15:50
  SPEAKER_00: Yeah, great question.
• 1:15:51
  SPEAKER_00: You know, it's so funny.
• 1:15:52
  SPEAKER_00: Um, for Parkinson's specifically, there's an app, I don't know if because of your your experience called Strive PD.
• 1:15:58
  SPEAKER_00: It's an app specific to Parkinson's, but it can interface with like your Apple Watch.
• 1:16:02
  SPEAKER_00: And I believe like a health system like Kaiser uses it for like a lot of the management of their patients.
• 1:16:07
  SPEAKER_00: I'm not 100% sure.
• 1:16:09
  SPEAKER_00: This is just what I've heard anecdotally.
• 1:16:10
  SPEAKER_00: Um, but it's so funny because patients will come to clinic, and I would say off the cuff, probably half of them, it's a very polarizing half are like, I track everything.
• 1:16:19
  SPEAKER_00: I can show you how much dyskinesia I had yesterday in these three hours, and other other patients are like, don't tell me to download an app, I will never do it.
• 1:16:27
  SPEAKER_00: And so, you know, it's sort of a pretty polarizing thing, but I think honestly, as a PT, it's fun for me because the folks who do track um, it's been great.
• 1:16:35
  SPEAKER_00: I think um any sort of wearables that track things like gate asymmetry, that's a good one now on Apple Watch.
• 1:16:42
  SPEAKER_00: Um even your indirect measurements of like VO2 max or um heart rate variability, sleep, really nice way to build in self-efficacy as a PT.
• 1:16:52
  SPEAKER_00: Like if I see them once every two weeks, we're up against all these signs and symptoms of advanced Parkinson's.
• 1:16:58
  SPEAKER_00: They can really do a lot and then come to me with their readout, and I can be like, that looks amazing.
• 1:17:04
  SPEAKER_00: Your step count increased by a thousand steps.
• 1:17:06
  SPEAKER_00: Here's what we're gonna do now.
• 1:17:07
  SPEAKER_00: So I think that helps for sure, but I would say it is pretty polarizing.
• 1:17:12
  SPEAKER_01: Yeah, yeah.
• 1:17:12
  SPEAKER_01: I think there's probably still a lot of um a lot of concern around, you know, uploading all health information to your phone and understandably things like that.
• 1:17:22
  SPEAKER_01: Um, because uh obviously there have been all sorts of horror stories in in terms of exoskeletons, like large largely speaking.
• 1:17:29
  SPEAKER_01: I know, I know it sounds like your primarily um your primary experience has been with skip, but um, do you find that people are embracing that form factor?
• 1:17:40
  SPEAKER_00: That is a great question.
• 1:17:41
  SPEAKER_00: Um I think it's too early to tell, at least in my experience.
• 1:17:46
  SPEAKER_00: I think people, like going back to your earlier question, people are really excited about any tech, especially those that freeze, because it's so tricky, whether it's sensors on the shoe, it's glasses, it's all the things, like those are the devices being made right now.
• 1:17:59
  SPEAKER_00: I think it's sort of like what needs to happen is like letting the dust settle a little bit, seeing what shakes out, and then really getting um devices on people longer term.
• 1:18:08
  SPEAKER_00: Like if they took it home, do they wear it day after day?
• 1:18:12
  SPEAKER_00: Is the cost, let's say, for the weight worth it in the long run?
• 1:18:16
  SPEAKER_00: Is the downing off time worth it because they're doing XYZ day after day after day?
• 1:18:21
  SPEAKER_00: You know, we just don't know.
• 1:18:23
  SPEAKER_00: You know, I to my understanding, it's still so early in the development where it's like things that are looking really promising in lab or shorter term, you know, that's where I'm curious about is in the longer term, you know, six months down, if the disease is changing on them, can the device be nimble enough to still help down the line?
• 1:18:40
  SPEAKER_00: You know, are they still choosing it?
• 1:18:42
  SPEAKER_00: I think that's where the data will kind of point us to your your answer.
• 1:18:46
  SPEAKER_00: And I don't know that we need to have it yet.
• 1:18:48
  SPEAKER_01: So we and we can close on this, um, but uh just generally speaking, like from a technological and or I guess scientific standpoint, like what what what excites you?
• 1:19:01
  SPEAKER_01: Like what what's being done in the Parkinson space right now that actually like gets you excited?
• 1:19:07
  SPEAKER_00: Yeah, I mean, I think first and foremost, the work that Skip is doing, I think I I wouldn't have continued to partner with them if I didn't truly believe in what they were doing, both how they're doing it and the product.
• 1:19:18
  SPEAKER_00: Um, I think in general, what excites me though is there's so many people that are looking at it from all different angles.
• 1:19:25
  SPEAKER_00: I mean, even products, like I was just alluding to, whether it's, you know, external queuing has such a big, you know, um body of literature supporting its effectiveness, both in Parkinson's and freezing of gates.
• 1:19:36
  SPEAKER_00: So people are putting, you know, cueing mechanisms on shoes, on your ankle and your socks, you know, all these things.
• 1:19:41
  SPEAKER_00: And then now starting to look at how does it adapt?
• 1:19:44
  SPEAKER_00: You know, can you do some sort of closed loop system to try to get at that instead of the body habituating to it and it becoming less effective?
• 1:19:52
  SPEAKER_00: You know, you talk about exos, you add things on to exos.
• 1:19:55
  SPEAKER_00: And I just think all of that's really cool.
• 1:19:58
  SPEAKER_00: I think combining that too with like exercise, I think there's so much great data to suggest, like, if we can get people a little more active, they're they're gonna be all the better off, you know, both quality of life, maintaining their life outside of Parkinson's and their community, and fitness-wise, we're getting less of that deconditioning and stuff we have to battle on top of the disease.
• 1:20:18
  SPEAKER_00: So I think that really excites me as a PT to get to work and see some of this come out and know I can contribute this exercise piece and things to further.
• 1:20:26
  SPEAKER_00: Um, and I think that's really exciting, you know, especially independent of where DBS goes and where medication goes and how that looks with it together.
• 1:20:34
  SPEAKER_00: Um, but I think the field is at a really exciting crossroads right now is more is being done in freezing of gate to define it and think about how we measure it, especially outside of the clinic.
• 1:20:43
  SPEAKER_00: Now you have all these devices coming in ready to sort of use that and run with it, which is cool.
• 1:20:47
  SPEAKER_01: Yeah, I mean it sounds like anything that is getting people like up and out is probably a net positive because you know, you for for you know, for reasons uh like personal.
• 1:21:00
  SPEAKER_01: My my my father who passed away recently had a lot of personal pride.
• 1:21:04
  SPEAKER_01: Um, you know, and and I think that that's probably something that maybe stops folks from like going out and then and then it becomes a vicious cycle, right?
• 1:21:11
  SPEAKER_01: You don't go out and then it kind of Yes.
• 1:21:14
  SPEAKER_01: Yeah.
• 1:21:15
  SPEAKER_00: 100%.
• 1:21:16
  SPEAKER_00: No, that's that's why I think as a PT, that's the thing that we see that we're just up against, and it's so brutal because we know the disease causes all these motor issues, and it also causes things like apathy and depression.
• 1:21:28
  SPEAKER_00: And so I think and sleep issues.
• 1:21:30
  SPEAKER_00: Like you don't feel like you want to exercise when you got one hour of sleep the night before, right?
• 1:21:34
  SPEAKER_00: Like, so it's just like we're up against so much.
• 1:21:37
  SPEAKER_00: But I think, you know, any way we can move the needle on, you know, again, getting the movement going, getting it, you're not freezing so much.
• 1:21:43
  SPEAKER_00: You can go back to the store, you can go to church, you know, those are a lot of wins that I think we start stacking in a way that really fuels everything else.
• 1:21:51
  SPEAKER_00: And so I do think, like I said, that's what excites me.
• 1:21:54
  SPEAKER_00: And I think um, I'm hopeful that in the next five years that we're gonna see a lot of these start to emerge where patients are choosing the thing, choosing the thing, going out with the thing consistently enough where we get to see that emerge.
• 1:22:06
  SPEAKER_01: Well, Jessica, thank you so much for taking the time.
• 1:22:08
  SPEAKER_00: Yeah, of course.
• 1:22:09
  SPEAKER_00: Thank you for having me.
• 1:22:10
  SPEAKER_01: Thank you so much for joining us today.
• 1:22:12
  SPEAKER_01: Thank you to Catherine and Jessica.
• 1:22:15
  SPEAKER_01: Thanks to you as always for tuning in.
• 1:22:17
  SPEAKER_01: If you've been enjoying the show, please like and subscribe wherever you get your podcasts.
• 1:22:22
  SPEAKER_01: And don't forget to subscribe to our weekly newsletter over at automated.fm.
• 1:22:27
  SPEAKER_01: We will see you the same time next week for another episode of Automated