This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.

 Gozio Health partners with health systems to increase consumer engagement using a proven mobile platform and strategy. The entire healthcare journey, both digital and in person, is improved by consolidating all patient facing digital solutions into one premium native mobile experience. Gozio customers surveyed in a class research emerging technology spotlight report were 100 percent satisfied and said they would buy Gozio again.

For more information, visit Gozio Health. dot com.

  Today on Keynote

(Intro)  If we're not working with those, that.

Are going to utilize the assets and the technology of what we're building, then what are we doing?

My name is Bill Russell. I'm a former CIO for a 16 hospital system and creator of This Week Health, where we are dedicated to transforming healthcare one connection at a time. Our keynote show is designed to share conference level value with you every week.

Now, let's jump right into the episode.

  (Main) Welcome to this week Health Community Town Hall Conversation. I'm Reed Stephan, CIO at St. Luke's Health System in Boise, Idaho. And I'm joined today by Jennifer Langton, who is the Senior Vice President of Health and Innovation at the National Football League. Jennifer, welcome. And thanks for making the time.

No, Reed, thank you for having me.

Now, I think our listeners generally will be familiar with the NFL. It's a well recognized brand and name, but maybe I don't think they'll know about your role, which is, I mean, the title is incredible. Can you just take a couple of minutes and maybe just share a bit about your career path, how you ended up in this role, and then at a high level, what does this role entail in terms of accountability?

Yeah, no, I appreciate that. Career initially led me to the realm of finance. I don't know if you had known that. And I was always looking at emerging technologies to restructure companies. And so prior to joining the NFL, I after a course of several years had restructured Atari, the gaming entertainment company.

And when I say emerging technologies, I had built all of their financial systems there. And we had redeemed growth and profitability. And so I always had seen things in my earlier career for emerging technologies and how they could enhance either processes or insights for companies. And that was finance.

While I was at Atari, I was named North American CFO. But there was something within me that I felt like I needed to go back to business school. So you can question stuff because my life is already very busy at that point in time. But I did go back to business school. But at that same time, an opportunity came for the NFL.

Where I was taking a role as VP of finance to restructure their apparel strategy, and today I'm VP of what I consider. One of the, in fact, the leading innovation platforms, leveraging AI and machine learning to better understand what players needs are as it relates to preventing injuries, recovering from injuries.

I grew up as an all American lacrosse player, and in my second year at the University of Virginia, I blew out my knee. One of the greatest challenges I had to overcome as an athlete is now inspiration for my career. So we had always. prefer to see our players. In fact, I would rather be on field than on sideline.

And so it does give me that inspiration. I've lived it. I've had to recover from injury and have to prepare for injury and then to optimize performance after having an ACL reconstructive surgery and then three surgeries after I did make it out on field. But it does give me that inner passion and fire to do what I do today.

And I love technology as well. Again, those emerging technologies and a better, more efficient way to do things is another interest of mine and passion. At the end of the day, it's pretty remarkable that what I do now in my career does have inspiration from different streams of my life.

We could have a whole conversation. Maybe this is a part two podcast on the importance of passion, fueling your career, but also learning about your career progression. You took ownership of it. You didn't abdicate your career development or your career opportunities to your employer.

You pursued what interested you and made it a point to discuss it. And I'll just leave that there, but that's a wise nugget you've shared with with our listeners. So thank you for that.

If I hadn't spent that time and really connected and positioned for myself at such an important inflection point in my life and my career, I don't think I would have been where I am today.

So I appreciate you, yeah commenting and asking that. But if you don't ask, you don't get.

That's right. And I think you just outlined the structure of the book you're going to write on kind of personal development. So it's beautiful. Okay. But I am really interested to dive into some of the topics you foreshadowed around AI and what the NFL is doing.

So you and I met a few weeks ago at a Amazon web services, healthcare advisory board, and you shared with a group. Something that was new to me, this digital athlete project that the NFL has undertaken. Take a few minutes and share with our listeners just what that is, how long you've been at it, and what the objectives are.

so the digital athlete, first and foremost, is a collaboration with AWS, you were in that meeting. That represents what I consider the next generation of player health and safety in the NFL. What the technology does and uses is artificial intelligence and machine learning to build the virtual representations of each player.

So think about a digital twin. We hear a lot about digital twins, whether for turbine engines, for humans, et cetera, digital twin. So that we could better understand what each player needs when they're training, in practice and engaged. So what we do is we capture in each three of those training practice in their environments, we capture a complete record of the player's data, with that data, what we're able to do now, this is years.

And I'll go through a bit of timeline and chronology, but what we are able to do is we can now build off of that player data, we can build a risk mitigation models. So that we can build those models and give them to the team so the teams can understand precisely what the players needs are as it relates to staying healthy, recovering quickly, and optimizing at their best.

And we're hopeful that this technology, we're not there yet. We're hopeful that this technology then will predict and prevent injuries in the future. And as I said, a team success is contingent on a healthy roster. And that's how important this is, the availability of healthy players. So if we do, and I know that we will, I'm convinced we'll get to prevent and predict injuries with the technology that we're still in development.

So that's just the overall of what the digital athlete is.

So that has to be, you think about all the teams in the NFL, all the players, and certainly the competitive, security you have to have around it for teams to trust that they can provide this data to the NFL. So you mentioned AWS, maybe just share what were some of the key capabilities that they offered that were crucial for this partnership and why you chose to use Amazon Web Services for this endeavor?

Yeah, I'm going to take a step back from that, right? When I started in health and safety we put together for a long time, the NFL gathered data that would analyze injury data manually. In 2016, we designed this engineering roadmap and the engineering roadmap's mission and goal was to better understand the mechanism of injury being concussion on field.

So that we can also transform the helmet marketplace. I use the analogy that a quarterback when rushed, he protects the ball and doesn't brace himself if hit. And so larger rate of concussion for quarterbacks is in the back of the head. Why wouldn't you have a position specific helmet that had more protective materials, computer science, material science problem, in the back of the head?

For that position. So what this engineering roadmap had done was, how do we better understand the mechanism of injury being concussion and help to translate the helmet marketplace when I say manual, we would review biomechanical engineers that were motion capture experts would review one concussion via video review and annotated hundred and fifty variables at the same time for the denominator of the research, we would take a collection of games and it would take four days.

And then you do a double blind review to get the denominator, meaning how many head impacts did happen in that game. So you can't do that at speed and scale.

Yeah.

Was it effective? Very effective. We really started to understand insights. And this is what you had opened with earlier. We started to insights to one, the helmets, speed, locations, velocity, where was that impact happening?

At what position? In addition, we started to see behaviors and plays that were leading to more injurious concussions, more injurious plays, more injurious behaviors. With this rich data set, we did have some early quick wins, but then we put a pause in it and said, what if we were to scale? What we didn't have was a cloud and cloud compute environment.

We had disparate data sources. I think about my, finance days read where it was like Excel, hell, everything's in Excel. So we were building a database, but the process was so manually that what we wanted was a partner and we were very clear. That understood our vision and in this audacious goal to predict injuries.

In addition, we wanted the best in class cloud and cloud compute. We also wanted a partner, not just a cloud and cloud compute technology that would help us better understand the immense amount of data that we were pulling in, not just data points themselves, but from video. What I just explained for concussion, when you're annotating what is happening in an injury.

In Excel, computer vision can now do that. So it was a partner that would help us to advance with technology, what we were doing manually to create data, not just the data that we were already capturing, right? And then also we wanted a partner that of course we can trust and work with. And so we had already had a very established relationship with AWS, the reason is they were a part of our next gen stats.

So it's a zebra technology that goes in your shoulder pads that measure coordinates on fields. And so you see those predictive plays, but we already had a well established relationship with them, but a very positive and healthy relationship with output. And that helps when you're forming then a partnership.

So the digital athlete was then introduced in 2019 and now in 2024, we'll get into a little bit in more detail. We are leaps and bounds advanced to not only the synchronization of those data, but the insights that we are reaming from models and from data that are making really profound changes to our game and helping the players as well.

Yeah, that's fascinating. I'm gonna just give some healthcare context because you're here talking to healthcare CIOs and Chief Digital Officers, and so the connection for me is, and I'm speaking generally, in healthcare, we are data rich, insight poor, you talk about Excel hell, not uncommon to find a health system or a hospital that Excel is their analytical tool of choice, and we have access to this incredible trove of data that's available to us.

And yet, really being able to tame and harness that and to glean insights. We talked about things like predictive analytics and we've got models to help predict sepsis, but it's so inconsistent and it results then in suboptimal patient outcomes it adds expense. Are there any insights you can share that you've NFL's project here that might translate and help Healthcare CIOs think through how they tackle this.

Data fusion for analytics with time elements, the synchronization of all those data points is key. Think about having one key element that ties whether it's a player or a specific to time. To build a complete 360 degree player view that I was referring to, what the digital athlete does is it gathers that vast amount of data like you have, right? You have a patient, you have your docs.

There's a tremendous amount of data. A few examples of that is one. From Gameday, the next gen stats, we have sensors in player shoulder pads that provide to us the player's real time location, speed, and acceleration and distance. That's in one time element. Second is we have sensors in RFID tech that are embedded in protective equipment so we can see what is worn every play and in practice.

We have an instrument, a mouth guard, that measures severity of impact. That is in a different time element in synchronization. How do you make sense of all of that data? Then you have video. And in video, you think we have video of every single game and practice provides an immense amount. of data. And what we have been able to do now is build computer vision and machine learning, like systems, they're algorithms that are able to track key events in the system, like impacts or blocks and tackles to give us a richer data set.

You don't have the common elements of time across all those data points. How are you providing the most rich insights? And so what we have been able to do, and this was years of really sort of cracking the code here is the digital athlete is now with those algorithms because of the data fusions with that time synchronization.

And it's not easy. But we had tremendous amount of resources over multiple years. But what we were able to do with that is those algorithms are now able to run. Millions of simulations of NFL games or specific in game scenarios so that we can see what players might be at higher risk.

We can run plays and model plays in a simulation format because you're not able to test those in game.

So the more

data that you have is synchronized, the more insights that you could do with it. So we can reconstruct conditions on how and when injuries occur. Without risk to a player. We would not have gotten there if it wasn't for that data fusion with the time synchronization, because of all that we capture on different data point frequencies and different time frequencies.

So that was the true key to get to where we are today.

I love that idea of simulation and that certainly would translate in healthcare how much better to simulate a patient's illness burden or some comorbidity and the outcome that might ensue from that rather than have to learn from the live experience with the patient.

So I think that's really compelling. So to what you just said, yeah, what are some of the innovative changes or adjustments that you have been able to pragmatically. implement based on what you've learned through the Digital Athlete Project.

Now that we can integrate and aggregate, all players, every team across 32. It's very different from what we are able to do from a league office perspective than what one team could

do.

So this year, what we rolled out was the digital athlete team portal.

And this technology was rolled out to all 32 clubs for the first time. And that portal really empowered medical and you'd say their training teams to work smarter, more efficiently. It's not like they didn't have. Athletic management programs and processes. They do, they are the most elite, but what they weren't able to do was benchmark their running back across all running backs across the league.

And you can imagine the modeling that we're able to do now when you can integrate and aggregate across all. And so the portal, when rolled out, what it includes is like daily training loads and risk mitigation models that I was referring to for the team, but across. The league's benchmarks, whether by position or whether by teams, et cetera.

And now, instead of individually. really managing or tracking workloads for a 60 man roster. This application on a day to day basis will give a team your five to ten players that are at higher risk with those loads and exertions because of the sophistication of the modeling. The more modeling you do, the more players, the more effective and efficient, and the more precise those models are.

And so we have moved leaps and bounds even in a year on the utility and the use of being able to integrate and aggregate that data, and I'm going to overly simplify the workloads of every single player, but with more data like you have with patient care, the better the learnings are, the richer the learnings are.

So we were extremely encouraged by the outcome over the course of the first year. We had a result that reduced, and it's not fully associated, but lower extremities in the pre season, and that was one of the key things that was new this year. Was the team portal being rolled out? We have a very high rate of injury in that first two weeks in the acclimation period.

And so we did see one of our first declines. What is really important though, with the team portal is the feedback coming from the teams. Certain models work, certain metrics work. Everybody thinks that they have the shining bell and whistle for one metric. That might work for one player, but not the other, but the feedback coming from them to continue to enhance and improve our models in our system.

And what is we. built it for them. We built it for the health and safety of our players. However, the teams, being the medical and the training staff, are the conduit to that player. And so the richer the insights and the information is to them, and then the feedback loop coming back to us is what helps us to continue to refine, enhance that application for their use.

  📍 Hi everyone, I'm Sarah Richardson, president of the 229 Executive Development Community at This Week Health. I'm thrilled to share some exciting news with you. I'm launching a new show on our conference channel called Flourish. In Flourish, we dive into captivating career origin stories, offering insights and inspiration to help you thrive in your own career journey.

Whether you're a health system employee in IT or a partner looking to understand the healthcare landscape better, Flourish has something valuable for you. It's all about gaining perspectives and finding motivation to flourish in your career. .

You can tune in on ThisWeekHealth. com or wherever you listen to podcasts. Stay curious, stay inspired, and keep flourishing. I can't wait for you to join us on this journey.

  Let me ask you about the human element of it around change management. From my simplistic view, you've got this league, you've got, The front office, you have the owners, you have the players, highly competitive, there's so much parity, like the slightest advantage could result in a win or a loss.

Did you have any challenge as you rolled out these capabilities? Did you have to overcome maybe reluctance or resistance by players or coaches or teams wanting to fully participate? And I ask that because from a healthcare standpoint, we have capabilities. We have computer vision, we have ambient listening, but there's a change management exercise.

There's clinicians who might feel reticent to utilize these because they feel like it's going to spy on them, or a patient who feels like there's a privacy concern. So I'm just trying to see if there's Threads that we can draw on and learn from to help us in our healthcare journey.

Yeah, no, there are a tremendous amount of learning in what you're honing in.

Here is behaviors are very hard to change. Like I said, athlete management systems, they have them They are training the most elite athletes. This has been more of a, let's say a golden gem on what they didn't have and the utility of it. I personally think that It is going to continue to accelerate leaps and bounds with getting them fully engaged.

But what helped us here, one of the key learnings, Reid, is that when we built the digital athlete in this team portal, We did not build it in an ivory tower. We didn't build it just with our biomechanical engineers. we build with a collection of pilot teams and clubs.

And so what that helps us to do is build more of that credibility, more of the learnings. It's that continuous feedback loop. They then become the ambassadors to other teams. What's very unique here is that the teams, and we show this. Internally, but the teams that have healthy rosters have been the Super Bowl winners in the past few years.

So the more information that you can provide them, it's education because they hadn't gotten it before, but it's when one is doing it well and is best in class, then it's like the others learn from it and continue. But I'd say a few of the key learnings is one, the pilot teams I'm continuing to build, and then subsets of the year we go back because we can't always get their full engagement in season, but subsets of the year is.

Getting that rich feedback from them. What more should we be doing? And not just from the teams that are pounding on the system, the teams that aren't, right? So I think there's like that constant feedback loop from them, and continuing to iterate that. And like you said, it's education. It is education.

You have to spend the time. We do one on one team calls. Every time we go to combine and we have our, medical committee meetings, we ensure that each one of like new staffs coming into clubs are educated on all that we are doing with health and safety. Big part of that, of course, is the digital athlete, but you have to take the opportunity.

This year. We did as well. Our first health and safety summit. We do an annual meeting with the owners every year. And this was a health and safety summit as well, because we want them to engage as a team so that the more that we're giving them. Medical or a team physician or your sports scientist is sharing what the digital athlete is with that team.

So it's broader education. As you can hear, there's multiple different, learnings based on insights. But I think that engaging. The user is key, helping them help to be that influence to either owners or different stakeholders is key because they build that then credibility and validity of the system as used.

They're not going to tell you their secret sauce, but, they are using it and what the learnings are. And there's always key learnings.

It's funny how despite the industry, there's just, there's crossover learning because at the end of the day, the work we do it's about humans.

And so what you just described, education engaging the stakeholders in the design highlighting successes, because then that will sometimes generate people that want to have similar results. And it made me think of last month, there was a one day strike by a group of nurses in California, and the way it was headlined, was nurses strike against ai.

But when you read like their letter, it wasn't about AI at all. Like they weren't saying we don't want to use it. They were saying, we want to be involved in how this is designed and how this is going to impact our day to day operations, which is completely legitimate. And so to your point, when you skip that step, when you think you can just design it, And then present to the users the finished product.

That's when you get in trouble. So love how you laid that out. And that's a key learning that we can't state often enough because we, in our haste to want to innovate, sometimes we skip that important step of engagement and you just can't.

Yeah, and always remember what your goal was, what your mission is to improve the health and safety for our players and our game,

to

be able to predict injury and keep your eye on that focus.

It's continuous iterations of new and novel technologies that have not been built, but you have to stay focused and you have to align that the stakeholders to help to do that.

Yeah, it's really

quite a comprehensive team.

Yeah, so you described this digital athlete project kicking off in 2019 and it sounds like it's really started to accelerate even more rapidly in the last year or two, based on probably a confluence of factors.

As you look to the future, what are you most excited about? What are the opportunities that you're most interested in pursuing?

I personally think that we had a banner year in health and safety, it's been our best year and why do I say that it's the. Analytics and the insights that we are able to glean from this system. When I was explaining that denominator of research with human eye, you're counting the number of head impacts.

We ran with AWS two crowdsourcing challenges, Reid. The first was to identify head impacts. You label the head impacts at the exact moment and that automated. The second crowdsource was the, how do you associate that impact with the player? So it's another algorithm. Now it's amazing what we can do.

Our NFL teams and our engineers and then AWS has quite a bit of, resources there to run that. But when you crowdsource these two algorithms, AWS and the NFL help us to solve for the end Result of that is last season. The data that we did have, if you can take those two algorithms, and so one labels the helmet, the other one with the impact and time, we'll synchronize that to the player.

The reason why we have the players, the next gen stats gives us the coordinates on field to locate that player. We have head impacts going back to 2015 for every player. When we're looking at injury mitigation efforts and putting strategies on how, we have been down and have sustained our reduction in concussion.

However, here, our focus is taking those egregious hits out of the game, and how do we reduce head impacts? This year, for the first time, we had rolled out with the team portal, and this was when I say behaviors. It's like we rolled out head impacts for linemen, offensive and defensive linemen, to their direct coaches, so they can put together strategies to reduce head impacts for the higher risk for the ones that have more impacts than others.

If you think about when we started this department in needs, you said health and safety, it's pretty unique and now health and innovation. It's we couldn't even count. And with human error, the number of impacts. And now we are furnishing our teams reports that are automatically counting head impacts to get the head out of the game.

So you can think of like how profound that change was with these type of models to this year, we made a change to our kickoff. And so if you think about kickoff, that was all the basis of the analytics for that work was done with the digital athlete. We could not have done that. The simulation modeling that I was talking about, we had a low rate of returns on kickoff was the importance and the excitement of the play.

But what our aim was, how do we increase the rate? Of kickoff, because it's exciting, but while minimizing injury, kickoff, speed and space, right? Grace, velocity and impact. And so the computer vision system technique, what we were able to do is identify those head impacts at each moment and create. A model that was for closing velocities will result in the number of impacts.

So it was an association. So we took our play in a simulation. We modeled it. We took the XFL's plays. We modeled it in that simulation and with. That risk mitigation for concussion on closing velocities. I'm overly simplifying, you can imagine, but we were able to see and adopt a modification of the XFL rule.

You can't test that kickoff.

Yeah.

We couldn't test it in game, all done by the analytics of the digital app. So it's pretty remarkable. One, we're giving head impacts information, and then I'll go to where we're going. The second is to be able to modify a data driven decision.

Yeah.

One of the key plays of our game, and have comfort and solitude that it's not more injurious when it was one of the most injurious plays, is pretty profound.

I mean, that's transformational. So that was completion, I would say, of this year. We clap and applaud, but move quickly on. So what's next? When I say predicting injuries, we're not there yet. And so predicting injuries is what we call pose estimation. The digital twin where we started, but if you can quantify player movement down to a level of joint, not just key point detection, then I do think, and this is right now in our R&D pipeline over the course of the next three years with AWS, but we would be able to better understand pose and postures.

What leads to injury and what doesn't so that we can help to predict them. So the quantification of that body movement. Not very easy is really what we're building on iterations right now. And it's to come, and then beyond, I mean, we talk about that. I was a lacrosse player, as I had mentioned, but you can imagine we're a helmeted sport.

We have so many inclusions and exclusions on our field, the complexity of it. But think about what our application could do for other athletes in other sports, members of the military, in healthcare,

We're so rich with data. We're so rich with research and the more that we can normalize and standardize that data.

to gain insights. I think that what we're doing with the digital athlete is that it really could have a profound impact, multiple different learnings on how we've structured the collaboration, how we've structured the work streams. Many people going back to my Excel days are always trying to still solve for getting out of Excel, right?

Somebody is trying to solve for that data fusion, how to use the data. Somebody has that key time element issue. And so we've been able to structure that to give us insights. And I do think that the rudiments of what we built for multiple different entities, industries will have an impact.

I think so as well.

And this is, you're saying that like my mind starts to go, we have an orthopedic surgeon who is passionate about working with high school athletes because his position is a lot of the injuries he treats are a result of bad posture, bad pose, bad form. And so he's trying to figure out how do I quantify that and then educate and instruct?

I think of our clinicians who oftentimes have injuries based on moving patients, which my assumption is a lot of that is based on maybe bad pose or bad form in doing so. So if we can somehow have the insight there and meet them with education or more awareness of what they're doing that can be done differently.

It's so much opportunity. So it's just, I'm just struck by Jennifer, you are a healthcare leader in your role, and I think you probably feel that, but it's just your passion and what you've done in these last five years and beyond is incredible. So I'll ask you this kind of open ended last question.

You've got this audience now of healthcare IT leaders. What words of advice or encouragement would you offer based on what you're doing and have done with the NFL?

One, as Reed, many times I'm the face of the work, whether for NFL or AWS, but it's always a team. I mean, I know all the algorithms, right?

I'm not a computer scientist, but

I can relate. Yes.

Yeah. These are really big. digital transformation and efforts that you do need quite a comprehensive team. I think that advice going back to the way in which we solve for concussion, not solve for, we made an impact right and reduce concussions based off the data that we were gathering.

Scopes of work like pilots, that are quick wins, that you're melding the team, the learnings, the insights, and grow from there is so key. People try to bite off these enormous scopes of work,

like the

team portal. Myself with engineers and others that are really provoking transformation and change. If we're not working with those, that.

Are going to utilize the assets and the technology of what we're building, then what are we doing? And so to make an impact, regardless of what department, what company or industry you're in, it's try the quick wins. Try the scopes of work to gel the team and the learnings, and then grow from there.

Because many times we try to bite off too much, and that's just learning. And then the behavioral changes, right? The cultural, data privacy. There's so much to learn that I feel like companies try to. There's a tremendous amount of sophistication, right? In what technology can do that people, I believe, are overwhelmed by.

Today, and we're trying to learning. So honing in on what can you sell for first and growing from there. I think we've gained a lot of success that way. And we built and scaled the team that way, partners that way. That would be my key advice. And again, it always is a team.

Yeah, I think that's beautiful.

Jennifer, thank you. This has been so fun. And the goal of this podcast is to just share insights, spark ideas and thoughts. And I have no doubt that you've accomplished that with listeners. So really appreciate your time. I know , this is not a mission accomplished moment. I don't think there

ever is such a thing, but congratulations. What you've outlined and described, especially the last five years, is really incredible, and just thank you for improving the health and safety of the communities that we collectively, I think, serve together at the end of the day.

I appreciate that.

Again it's a robust team, but we have some of the really is excellence in from a biomechanics to data scientists, et cetera. So I really do appreciate that. But it is very rewarding.  

Thanks for listening to this week's keynote. If you found value, share it with a peer. It's a great chance to discuss and in some cases start a mentoring relationship. One way you can support the show is to subscribe and leave us a rating. it if you could do that.

Thanks for listening. That's all for now..