The below is a transcription of the podcast BOOM: Biomechanics on our Minds, Episode 14: MAP BOOM and the Future of Wearables. The episode features an interview with IMeasureU co-founder Thor Besier. Thor also founded Stanford University’s Human Performance Lab and is currently an associate professor at Auckland Bioengineering Institute.
Click the play button to listen or read below.
Melissa: Welcome to biomechanics on our minds. Welcome to BOOM.
Hannah: I’m Hannah O’Day.
Melissa: And I’m Melissa Boswell
Hannah: And we’re here to talk about some fun biomechanics.
Melissa: Yes, we are. So, this we’re grad students at Stanford University but this podcast is brought to you by the International Society of Biomechanics. And this episode is brought to you by Advancing Women in Biomechanics, because you should go to the advancing women in biomechanics event at ISB ASB 2019 in Calgary. Yeah, so the event is on August 3rd at 7:30 p.m. And it’s for everyone interested in helping advanced women in biomechanics. So, men and women we want everyone there to come and join the conversation, share experiences and learn more about the challenges that are faced by women in our field and discuss some proactive solutions.
So, that’s the Advancing Women in Biomechanics event and you can also follow advancing women in biomechanics on twitter, at AW Biomechanics.
Hannah: I’m gonna do that right now AW, oh wow it even comes up when you just put in AW, so it’s easy.
Melissa: There you have it, it’s easy to do.
Hannah: Even a caveman could do it.
Melissa: Alright, so today we will start off with a bit of boom and then we have an interview with Thor Besier from Auckland Bioengineering Institute in New Zealand. And we talked to him about wearables and their applications.
Hannah: Yeah, that’s super fun and Thor is a really fun guy to talk to. But first a bit of boom. So, today’s bit of boom is brought to you by the blog Biomechanics in the Wild, which is hosted by Maria Holland. And she is at the University of Notre Dame and used to be here at Stanford. You can check out her blog, the link is in the description of this episode.
Melissa: Or search biomechanics in the wild.
Hannah: Ok, it’s the second one when you google search biomechanics in the wild, because biomechanics is everywhere.
Melissa: So, we’re gonna be talking about one of her blog posts which fact are you gonna be sharing with us.
Hannah: Today well this one just struck me, the title is Can We 3d Print Our Own Skin?
Melissa: I am shook, tell me more.
Hannah: So, yeah this sounds kind of crazy and even seems like something out of a science fiction movie. But a researcher at Graz University of Technology, the Wake Forest School of Medicine and the Universidad Carlos III de Madrid have been working on creating an artificial skin that can sense temperature humidity and pressure. And the cool thing about it is that currently artificial skins usually are only able to measure one sense at best. But these researchers are using nanoscale sensors and developing them so that they can do all 3 senses at once.
Essentially they are, it’s a combination of a smart polymer core and a piezoelectric shell. This combination can take advantage of the different capabilities of the two different materials that are being used. The smart polymer core can detect humidity and temperature by expanding. And the piezoelectric shell actually can detect pressure through an electrical signal, that’s created when pressure is applied. We have prosthetics for vision, for hearing, but it’s crazy to imagine sort of this prosthetic for a sense of touch.
What’s even more amazing? I think is that they’ve made a handheld 3D printer that produces human skin.
Melissa: That’s so nuts.
Hannah: Yeah just let that like I’m getting like flashbacks to Silence of the Lambs. I don’t know if anyone has ever seen like there’s crazy and essentially has a different way of getting human skin.
Melissa: So, it’s able to 3D print like this combination of different materials.
Hannah: It can essentially create different types of skin cells and you can apply these layers of synthetic skin or printed skin directly to wounds. So, they could replace skin grafts you know when skin and a burn or something like that or some kind of trauma. They’re able to actually just print the skin. Specifically, they can cater it to which layers of tissue have been disrupted by scanning the wound.
Melissa: Very cool.
Hannah: Yeah, that’s it’s really crazy.
Melissa: I wonder how far they are from having something that’s actually usable.
Hannah: Well, I think that it sounds like they’re close as far as research and technology but they haven’t yet been able to test it on people.
Hannah: But apparently they’re being used in industries such as with L’Oréal to limit testing on humans and animals. They can test lotions and products on this fake skin or on this printed skin. So, that’s pretty cool and apparently, they’re being used in robots. And this blog post on biomechanics in the wild has a really nice YouTube video about a robot with this 3d printed skin.
Melissa: All right thanks that’s super interesting. And now for our interview with Thor.
With us today is Thor Bezier who is an associate professor at the Auckland Bioengineering Institute in New Zealand and also has a joint appointment with the department of engineering science. Thor, you completed your PhD in musculoskeletal biomechanics at the University of Western Australia and then you came to Stanford for your postdoc and actually set up the Stanford Human Performance Lab, where we work now. So, thank you for that.
Hannah: Thank you for the bean-shaped desks.
Thor: You’re welcome, yeah they’re pretty trendy. Aren’t they?
Hannah: They are really trendy.
Thor: Maybe they were when I put them in but that was 10 years ago, so maybe not quite as trendy now.
Melissa: Thanks for speaking with us today. So, your research combines medical imaging with computational modeling and looks at mechanisms of musculoskeletal injury and disease. And in your bio, it talks about some open source modeling tools to generate musculoskeletal models rapidly and then diagnose and treat injury and disease. So, could you maybe elaborate on this a little bit more some of the open source modeling tools you use and some of the other tools that you use to measure musculoskeletal parameters or things that you look at?
Thor: Sure, yeah when I first came to Stanford, I had an interest in creating musculoskeletal models from medical imaging. And so previously I’ve been exposed to using Sim back then there was no such thing as OpenSim, but I’d be exposed to musculoskeletal modeling and had looked at ways of estimating muscle force from EMG which is what I did in my PhD with David Lloyd in Western Australia.
But we really wanted to get into more detail and understand you know what’s happening at the joint level. And to do that we really then needed to look at creating models from medical images and this is kind of I think we’re the field of biomechanics is headed in general and it’s trying to really catch some inherent physics and physiology of biological systems. And to do that well, I think that the more we get into this and create complex models we realize that our models really need to best capture the anatomy and particularly when we’re interested in joints and joint mechanics.
So, when I arrived at Stanford you know spent a lot of time segmenting MRIs and looking at medical images and then said well, you know a lot of time and effort is spent by a lot of us all around the world making these medical well making models for medical images and surely there’s got to be a better way of doing this. So, when I came back to New Zealand about eight years ago I came to the Auckland Bio-engineering Institute and I’ve been well known for many years in developing methods to scale and fit meshes to cardiac, most mostly cardiac models.
So, they’ve done a lot of work on modeling the heart, and some work also had previously been done modeling the musculoskeletal tissues. So, I just basically said look why don’t we try and make these tools more available to other people, because I saw an opportunity to make these tools to help people segment and create kind of accurate musculoskeletal models.
Melissa: – Focusing more on the transfer of medical images to the models, like that pipeline?
Thor: Correct and kind of the fitting of models and for the OpenSim community who typically take some information from motion capture markers and use that to kind of linearly scale some bones. This is taking it one step further and saying well what if you had some segmented data of those bones as well as the points in space and you could do a better job of maybe getting customized bones and joints.
So, that’s when we created this project and it’s called the musculoskeletal project, and I thought it was you had a clever little acronym, ‘MAP’ of course. And don’t laugh it took me a long time to think of it.
Hannah: It’s like us trying to come up with ‘BOOM’.
Thor: Yeah. ‘BOOM’ didn’t quite fit with the acronym but maybe we could come up with something that was met boom. So, we said about then creating a set of tools which are all in Python and hence kind of open-source. So, we wanted to make these tools available to everyone where we were then you know rapidly create musculoskeletal models and that was some time ago, now we’ve done some of that work was funded by FDA, USFDA. And you think why on earth are the USFDA funding these guys in New Zealand to make some kind of modeling tools.
But they are critically interested in using computational models for regulatory science, and so we have an opportunity then to create an open-source suite of tools to generate populations of models and test medical devices in a kind of general sense on these models. And some of the databases then we use a kind of an anatomical and functional data, but one of the key parts to this is an anatomical atlas that we get from post-mortem CT, actually from Victoria and Melbourne Australia.
And so they have database of around 60,000 full-body CTS and we have just a subset of that, about a thousand. But from those thousand CTS we can then generate models and the idea is that you have hundreds of models put per decade of life and we can start interrogating these models for really interesting kind of form-function relationships. But in the background we can now generate, rapidly generate, musculoskeletal models of lots of different individuals much more subject specific, but also anatomically detailed than what we could before.
And this year we hope to then complete that dataset in a way that we can then creates tools for the OpenSim community to rapidly generate models. And that’s kind of a very long-winded way of saying that we’ve created an atlas for people and it’s open source.
Melissa: Okay, so can you give a description or overview of like what the flow would be like? So, you talked about testing surgical instruments on the models or like what would be an example of something useful?
Thor: A pretty good example might be in orthopedics. So, if you’re wanting to design a new implant for example, and this is maybe a femoral component. You would like to know then what the variation is across the population of this shape and size of the femur. So, from our population models we can see the outer size but also the inner cortical thickness of the femur, because that’s all inherently built into the models. So, you could then test out that implant and say well this is how it might fit across the population, and you can start then narrowing down different demographics and say well this might fit really well with this demographic.
Maybe this fits well with men but not with women and you can do these types of analyses or of course in silico instead of doing the traditional approach which is you know putting these limited number of implants into some cadaver specimens and saying, look this is how they work. And so that’s kind of a nice example. We also have some ongoing work which I’m a part of which is an NIH funded project with Amit Adhamiya, who’s at the Cleveland Clinic. And that’s also with colleagues at Cleveland State and Denver University and the Hospital for Special Surgery in New York, and there’s five groups of us who are trying to develop computational models of the knee and in particular, we are interested in the reproducibility of the results.
So, it’s kind of like the grandeur challenge that we did with BJ Freglee and Daryl de Lima some years ago, where you kind of have some ground truth that you try and test your models on. And in this case we’re interested in joint mechanics, so we have some data from Denver and Cleveland where they have some cadavers and robots and our models try and predict then the type of behavior that you can measure then in those experiments.
And by doing that we can give, potentially we can give to the FDA some idea about what kind of reproducibility you might expect with these computational models which then could be used back to the example which could be used by an orthopedic implant company to say virtually test their design and say how well does it fit in the population.
Melissa: That’s awesome.
Hannah: Well, that’s amazing and yeah I like that, you really highlighted that you’re such an expert in modeling, but still needing to be able to validate and do things experimentally with that ground truth data, it makes a model valid.
Thor: I think it’s kind of an interesting point right, there’s a lot of people who do musculoskeletal modeling or computational modeling in, they’re kind of an arm’s length from the experimental data collection. And until you actually do experimental data collection yourself, you don’t quite appreciate what goes into it, whether that’s with you know even Categoric tissue or whether you’re dealing with human subjects. But I mean there’s a lot to it that goes into collecting those data but of course you know a model is only as good as what it can predict. And so those experimental measurements are critical for us if we want to bring our musculoskeletal models you know into a clinical setting.
We need to be able to say you look these model under these conditions, these models can reproduce the results that you might measure experimentally.
Hannah: I guess on that note, I see you’re also doing a lot of work now with inertial measurement units or IMUs and wearables. And that tends to be a field that’s it’s not new but it probably seems to be less validated than say traditional motion capture or ways that we’ve been measuring joint kinematics in the past, in a laboratory setting. And now we’re going out into you know the outside world where things are a lot more variable and we don’t necessarily have that really awesome gold standard data and ground truth.
Thor: Yeah, it’s an interesting problem really, because at one stage we want to rush out into the real world and start measuring people in the real world live and you know a colleague of ours at Harvard, Irene Davis, you know this is biomechanics in the wild. And the challenge, of course, is then having enough fidelity to measure what you really want to measure and in most conditions. And where it’s interesting is where you know you see some wearable technologies that have been developed and it looks like cool and interesting technology. But whether or not it actually gives you the information you really want with the fidelity that you need is questionable.
And there’s been some really nice work done on this in the past, Ken Kaufmann for one at Mayo Clinic’s done some really nice work, comparing different commercial sensors and under different conditions. You know it turns out some work under certain conditions really well and others don’t and there is no one seat there really that works well in all conditions and it’s really. And that’s kind of to be expected for people who do computational modeling, again our models are tested under certain situations and circumstances and we validate them for want of a better word under those conditions.
So, sometimes if we perturb them or test them under conditions they’re not designed for, they give undesirable results. So, that’s why I think it’s an interesting area because we need to be careful about what we measure and how we interpret that information.
Melissa: Right, as it seems like with motion capture and other methods in the past, we’ve been focusing a lot on joint kinematics and like joint angles and what effect those have on the musculoskeletal system. But now with IMUs your motion, mostly capturing like accelerations and then the angular velocities, and it’s more difficult to get to the joint angles. So it’s a, I guess my question is if you think that just maybe using like raw data, how we can then figure out what this new type of data means for the musculoskeletal system. Like is there another way to go about this than try to calculate the same measurements we’ve been using before?
Thor: You’ll in some sense, it’s quite nice that you can measure the accelerations and you know if you know computational modeling methods. Sometimes those accelerations are exactly what we want and traditionally we double differentiate our displacement data from motion capture to get to those accelerations. So, in some sense of measuring those linear accelerations directly with IMUs is quite beneficial it gives you some insight. Of course, the challenge comes if you want to integrate that information into displacements and that’s where things start to kind of fly apart literally.
And you know there’s a couple of approaches, some people will do brute force kind of machine learning style approaches with these sensors. They’re just collecting volumes of data and then letting the model or some ground truth, provide a black box interpretation of what’s going on and that’s been shown to work well. Again under certain circumstances and constrained conditions, you can train a model and it can reproduce data quite well, quite accurately and in some cases in real time.
A different approach is to have more of a mechanical or deterministic type model which is saying, well if we have these sensors placed on these body parts and we have a physics-based model as a constraint, then hopefully we can match these two things together. And there’s only so many ways, once you add the constraints of the body, there’s only so many ways in which you can be moving to give rise to these linear accelerations and angular velocities that you measure. And that’s kind of an approach that we’re investigating along with BJ Frigley, who’s at Rush University we have a new initiative grant to look at, can we actually do this in real time and provide new tools for the clinic in terms of gait retraining and monitoring people’s gait.
Hannah: That’s awesome. So, could you tell us I was gonna ask you actually if there was a project or something you’ve been able to accomplish with these IMUs that you wouldn’t have been able to do before with mocap or you know the traditional set of tools that we’ve had in the past. Is there some component of that’s possible only because you’re using wearable sensors versus the traditional gait lab?
Thor: Yeah, I think this gets to the point really of where traditionally we’ve measured people in very constrained environments, and look this has represented of these five steps that I’ve measured which are the good you know force plate strikes that you get in the gate lab. A representative of what you do you know as in terms of walking when you get outside, and we know that’s just not the case. And so this is then an opportunity to say, well firstly can we measure gait in different environments with enough fidelity that we can get a reasonable estimation of what the joint kinematics are, but then more interestingly monitor over time.
And so another project that we’ve got, I mentioned Irene Davis so we’ve got a project now that’s funded by GE and the NBA and we’re interested in tracking basketballers throughout their training and their season. And so this kind of data you would never dream of being able to measure previously without wearable sensors, but now we can instrument these basketballers, and we can estimate impact loads with every step they take.
So, that’s an amazing opportunity, but it also creates challenges because you have, as you can appreciate, a heap of data and you say, well how do we interpret those data. That’s where I think the real interesting part of this is, not just necessarily in say machine learning approaches in AI which is you know bandied around, it’s a term that gets used a lot. But I think if you have a valuable computational model which can help interpret those data and also have some fundamental understanding of the biophysics, and that’s really where I guess the mechanobiology comes in where you understand that the loads and how these loads influence the biology. Then you can get some interesting interpretation of those data that you know we could never do with these types of studies before.
I mean Irene to the study a couple of years ago and she tracked, I think over a hundred runners during the Boston Marathon. And she captured you know every step that they took throughout an entire marathon, which is just not you know possible without wearable sensors.
Melissa: How do the NBA athletes feel about wearing these sensors? Do you know their engagement with being monitored?
Thor: Yeah so it’s actually an interesting point because we’re (IMeasureU) having to go through the NBA Players Association right now to become certified as a sensor that we can actually use. We have to certify the sensors with the NBA, because currently the wearable sensors are pretty new and the NBA are very hesitant, perhaps as they should be, of monitoring things like, we call it load for want of a better term, but you know we’re trying to get these biomechanical measurements on these players. And the reason why they’re, the hesitant about it is that could be used against them.
You can imagine a high profile player who potentially could be at risk of injury, maybe who has these data collected on them could be used against them really to say, well you know this plenty of players who would get paid lots of money to sit on the bench because of injury. So, the NBA are a little cautious, perhaps as they should be, with the use of wearable sensors. Now having said that, so we’ve collected lots of data on the New Zealand Breakers side, which I’m sure is well known to US listeners. But that’s the New Zealand basketball team that competes in the Australian League and we’ve got two years of game and practice data with the Breakers.
And so I think once we educate the players and say, look what we’re trying to do is characterize the training that you receive and the types of impact loads that you receive during game and use that really to help in your training and help try to prevent injury. Then I think the players get some buy-in on that but you know initially I think, they’re wary of the sensors and what you’re measuring what you’re trying to do.
Melissa: I saw a presentation that you gave on, IMUs in the military context, that was pretty interesting and I was wondering if you could elaborate on that. If you’re working on that study and maybe talk about that a little bit more.
Thor: Yeah a couple of applications here, one of the obvious ones really that comes to mind is the high rate of musculoskeletal injury in early recruits. And this, you know, on the outset it probably makes sense right, you look at this and you say, okay you have a whole bunch of young recruits who come into this program and you put them through boot camp and you’ve got to achieve a certain amount of physical activity. And you are taking people from a certain level of load and load history in terms of their musculoskeletal system, and you’re putting them through their paces and trying to get them up to a certain level of fitness, that obviously involves lots of stresses to the both the physiological system, but also if you thinking about the biomechanics you know the loads and conditioning of the musculoskeletal system. And so there’s really a mismatch between the load and load history that you have going into training and what you’re exposed to. And it’s no wonder that you know 20% of these recruits get some sort of musculoskeletal injury, whether that’s kind of a repetitive stress injury or bone stress injury which are pretty common or you know muscle tendon, tendon problems.
And so for us, we can say, well if we can manage that load better by understanding what your load exposure is, then we could potentially bring you up to these levels without necessarily pushing you over the limit in it and having too much load which causes injury. The other application here is understanding the context of these individuals in real-world settings. And so often people would like to know what actually happens out in the real world and unless you are kind of tracking people with monitoring systems, sometimes GPS works okay for monitoring where people are.
But it doesn’t necessarily give you the context of what they’re doing at a time, and IMU’s use can provide some information to kind of fill in the gaps. So, as a couple of examples there where (IMeasureU has) done a little bit of work in the military context. But for me the really interesting thing is to really understand the loads and boundary conditions really that are exposed to the musculoskeletal system and the potential for injury, even if you don’t manage those loads.
Hannah: And has the military been, I mean you talk about the NBA being wary of having that data available or at least you know being used for research. How is the military with sharing things like that?
Thor: Yeah, I think they’re quite different, I think at least we’ve worked with, we’ve got some projects and with different groups. But I think they’re pretty open to the idea that you can measure and monitor and by doing so you can gather information and data that otherwise you would be blind to. You just wouldn’t know if you have enough or too much load and the traditional training method really has shown to have a high incidence of injury. So, from that standpoint and I think any information they get is useful information.
So, I think they’ve far more likely to take it on board but it does bring it another interesting point which I haven’t mentioned yet. And that’s the concept of embedded sensors that you don’t even know were there. And so this could be incredibly useful for things like first responders as well as you know military but even sporting applications. And we’re seeing, of course, the miniaturization of these sensors that it’s you know it could be quite possible of course to embed the sensors into shoes and you use inductive power to you know charge up those devices and you don’t even know they’re there.
So, it becomes less of a problem in terms of compliance.
Hannah: Wow, the future is bright I think for us biomechanists.
Thor: Yeah, we’ve got a lot of work to do. I agree it’s never been a better time I think to choose bioengineering or biomechanics. I think as a form of study and I tell this to all of my undergraduates who I teach here and in bioengineering that they’ve picked a good time to step into the field. I think it’s really exciting.
Melissa: What are you most excited about for the future of biomechanics, even like within this area or any area of biomechanics?
Thor: Well, I think in general what the rest of the world is kind of understanding now and this is true for biology, it’s true for medicine is the understanding the importance of the mechanical environment. And you know we’re exposed, our tissues are exposed to mechanical forces and those mechanical forces regulate the maintenance of tissue. And they’re also important obviously for understanding injury and disease processes. So, more and more, the rest of the world is understanding the need to understand these mechanical loads.
And so our role I think in the future here is really to understand that a personalized level, what and in medicine is moving away from these cohort evidence-based kinds of studies which treats everyone essentially as kind of a homogeneous group. And say, oh let’s do an intervention on this group and see how they do. They’re getting away from there as they should to treating an individual saying, look let’s treat everybody as a person. And I think that’s you know that’s a space that we need to be in and we’re obviously pushing that in the medical field.
But as I said in understanding general biology, interesting biological questions, more and more we understand the need to personalize and the mechanical loads that regulate a lot of these processes and critically important. So, you know we’re sitting in an interesting space and I think there’s lots of different fields that that will rely on our expertise to be able to pull all this together.
Hannah: Yeah, I think that’s like a huge point of moving towards like personalized and individualized medicine, treatment, therapy, whatever regime even just like understanding, an understanding level of. But I feel like it’s also, it’s kind of an oxymoron. It’s like we talked about needing lots of all of us you need like you know a high number of subjects and things to validate what you’re doing and validate any methods you develop or the tools you’re using.
But then at the same time, you want some sort of specificity and you know ability to have like high-resolution data that you can pick out these individual differences and make more personalized decisions. So, I feel like it’s always balancing those two things kind of like what you said.
Thor: We’ve definitely felt that you know in the past 20 years that I’ve been doing this type of research that you know we get to this point where our models become a bit more personalized, a bit more complex. And it feels like you know where are we going to stop, which level can we just back off and say, well we’ve got enough information now that we can make some clinical judgment. And it probably depends on what you’re trying to intuit or interpret from your model. But you’re right that it seems a bit odd that you know we’re investigating, spending all this time and effort to do this at a personalized level and ideally you’d like to then back off a little bit and say well what can we generalize about these results to make some interpretation, make some sense of the underlying biology and what’s going on.
And I think you know we have to deal with those issues and it depends on the research question of course that you’re asking. But I think more and more and we’ve found this from the grand new challenge at least you know you need a certain amount of complexity to do a good job at predicting individual you know joint forces at least from that example. You know the average generic model that’s kind of scaled to an individual just doesn’t seem to do a good enough job. And so you know depending on what you’re trying to do, you really need some more level of input from the individual.
Melissa: That is really awesome, thank you for talking to us about that. We have one more question to ask you. And that’s if you have a fun fail, research fail that you would be willing to share with us?
Thor: Well, I probably have more than the average person in terms of failure. You know I think there’s been so many times where you expect to see something and of course it doesn’t work out. And whether or not that’s a good or a bad thing I think you just have to accept that it is what it is, and sometimes you just have to go with it and report it. And oftentimes, I think people are so sure when their hypotheses and ideas about what it is that they want to show, that they’re willing to gather all kinds of data and information they have to really support their hypothesis.
But I think at some point you have to be good, I think diligent enough to sit back and look at it and say, well you know what I’ve tried this the data doesn’t seem to support what I’m saying and back away from it and come up with something else. It may not be exactly the type of answer you expect from that question. But I think it’s an interesting insight into and when you look particularly back at a lot of different paths of people in their careers and what they take and they set off on a journey to try and solve a problem.
But we found this definitely in our now patellofemoral research at Stanford, our initial ideas and concepts just didn’t seem to add up with the data that we are collecting and we had to kind of reassess what was going on. And so I’m not saying that was a failure on anyone’s part, but we definitely learned and we’re humble enough to be able to reassess the situation and perhaps say, well maybe our intuition wasn’t right at this stage and let’s go out again with some different questions.
Melissa: I think that’s a great point, I think sometimes you can get so attached to your work and the same questions that you’ve been asking. But be able to take a step back and really remember your motivation and like big-picture for the work. And like at the end of the day we want to be able to help people and be able to advance in this field of medicine. And so I think that’s a really good point to make sure that your work is still asking the right questions.
Thor: Yeah and be brave enough to critique your own work right, that’s probably part of science.
Melissa: Yeah absolutely. Well, actually there’s one more question that I want to ask that’s probably the most important. If Hannah and I come to New Zealand, will you teach us to surf?
Hannah: Yes please.
Thor: Of course everybody’s invited.
Thor: You know it’s a direct flight from California, it’s easy. It’s only 12 hours.
Hannah: We could do surfing IMU study.
Thor: Yeah, well and it’s still nice and warm here at the moment, so you guys better come down before it gets before winter arrives. But you’re welcome any time.
Melissa: In 12 hours.
Thor: See you soon.
Melissa: Well thank you Thor.
Hannah: Yeah thank you, it’s great to meet you.
Thor: Yeah, well thanks for having me on the podcast. I really enjoyed the work, keep it up. Do I get to say boom now?
Melissa: Yeah, you get to say boom. Do you wanna countdown?
Thor: Yeah, I think we better.
Melissa: Okay, one, two, three.
Hannah: Melissa, this is my favorite part of the episode and I hear that you have a fail to share.
Melissa: I do, I totally have a fail. So, recently I was trying to synchronize cameras with motion capture like synchronize 2D cameras and motion capture cameras in the lab. And so I bought these really fancy pants cameras to sync with it, and they didn’t come with lenses right. And I’ve never, I’m not a photographer, so I didn’t know that you could even buy a camera without lenses like I didn’t understand that.
I mean I’m not like the really big ones, but these are just like tiny cameras. So, I had to find lenses for it and so I found lenses that fit the camera and so I bought them. And then I was trying to get the cameras to work. So, I plugged him in and everything was really blurry and I just could not figure out what would be the right setting to focus on something. And it was super frustrating and I went to go take the lens off and when my hand was like two millimeters from the lens, everything went into focus.
And I realized I bought a camera who’s like focal range was like eight millimeters or the lens’s focal range was like eight millimeters. Yeah, which is not ideal for recording someone walking.
Hannah: I mean maybe if it’s like ant-man walking.
Melissa: Perhaps yeah, so I had to get new lenses and I finally got them. And its actually just kind of been a mess because the videos were like three gigabytes for recording for like three seconds, which is just hot mess but anyway that’s a fail for another day.
Hannah: Well thanks Melissa for sharing.
Hannah: And that’s a great psa to everyone who needs their lenses. You know might need to know that their cameras won’t have lenses.
Melissa: Yeah and things to think about when you’re buying lenses for your camera.
Hannah: Oh I have its sort of, it’s a fail but it’s like weird. Basically I have a Mac and I upgraded to Mojave like the new operating system. And I’m always like hesitant to upgrade to things because I’m like uh like -.
Melissa: Yeah you never know have on the other side of that upgrade.
Hannah: Right, you never know and you can’t go back which is the scary part. So, but you know I was sick of the window popping up like five thousand times a day saying upgrade. So, I just like fine and I have in to it and I like closed everything I was like, okay I backed it up blah-blah. And I upgrade and then I go to open illustrator to like help to like edit a figure. And it’s just like it won’t work and I was like what, and then so turns out we use like illustrator on like a license.
It’s like a department license, so I have to like do some crazy things so like VPN in. I don’t know there’s things that words I don’t understand but things happen.
Melissa: Yeah things happen.
Hannah: And apparently it wasn’t illustrator that wasn’t opening, it was the software I needed to like access it, wasn’t opening on this new operating system. And I’m trying, I’m trying, I’m trying and then like I like try to do all these different things so trick it to let me like open the software but it won’t. And so I asked one of our lab mates and I was like and he’s like let me just Google it and I was like no I googled it like of course, I’ve already googled it and like tried the things that they say to do.
Hannah: And he finds this random post by this person, that’s just like download the software again. But you have to download it like three times and open it the third time. So, open the third downloaded like you know it’s like you download it once, don’t open it download it twice, don’t open it. It won’t work then download it the third time and if you click on the third downloaded one.
Melissa: That cannot be real.
Hannah: It opens.
Melissa: That’s amazing.
Hannah: So, for all of you out there that might need to just use something, now forever if something says it’s not gonna open and try it three times.
Melissa: Is it third time’s a charm, that’s what they say.
Thank you for listening to this month’s episode of BOOM. Thank you to tour for a great interview and you can follow the International Society of Bio mechanics on Facebook and on Twitter. How do you literally do this, I can’t say it? And on Twitter at IS biomechanics.
Both: Biomechanics off our minds
Okay actually should we introduce you like Professor Thor Besier?
Speaker: Yeah I’m actually not even a professor.
Speaker: I’m just trying.
Hannah: I’m trying professor.
Speaker: I’m trying professor. I’m an associate professor, my official title.