I loved Huey, hearing about Huey. That was so fun, and to see everything that that can do and all the predictive analytics in that. I think it's a very exciting opportunity to enhance our ability to deliver better care and reduce some of the burden we have on certain, so the human repetitive tasks. Yeah. I'm excited to see it in action. I think it's got a lot of potential. I like the positioning as being a supplement to the staff and the processes we have versus a replacement. I think we've got a lot to learn about it, so I'm excited to see what it can do. You have over a hundred thousand responses a year from your patients. We wanna utilize that and leverage it, and I think Huey and the AI technology is gonna be able to do that for UCI. Huey was my favorite. I took their mascot. I'm very excited. This morning, I was jumping practically up and down. It's groundbreaking. It's mind-blowing. It's a whole game changer. Hello, and thank you for joining us for NRC Health's webcast today, AI Meets Empathy: The Power of Human Understanding in the Future of Healthcare. We are excited today to lead this panel discussion. My name's Emily Adamson. I'm our content manager here at NRC Health. Excited to interview our experts in just a few minutes. But first, I wanna point out some key features in our webcast platform today. First, take a look at that Related Resources box. You'll find lots of additional content on today's topic: articles, podcasts, videos, and a lot of other things we think you'll find interesting. I also wanna point out the attendee chat. It looks like some of you are there already. It's a great place to continue the discussion with your peers and colleagues across the country today, and if you haven't already, please feel free to hop in and let us know where you're joining from. Finally, my personal favorite is the Q&A box. While I have some questions outlined for today's interview, I would love to work in your questions as well. So as those come to mind, please include them there, and we'll be sure to work in as many of those questions as we can. Today, we are going to be talking about the impacts of AI in healthcare experience management, and really how to bridge that gap between technology and human connection across the healthcare experience. This includes patients, families, consumers, communities, and of course, our healthcare workforce. NRC Health recently launched our own AI, Huey, healthcare's first AI engine built exclusively for experience management. During today's interview, we're really gonna explore the use of AI and next gen tools to improve experiences and keep healthcare human. I'd like to introduce you now to our panel of NRC Health experts. You can read their bios in the speaker box on your screen. We have Jennifer Baron, Chief Experience Officer here at NRC Health, Christophe Louvion, Chief Product and Technology Officer, and Vinitha Ramnathan, Chief Product Officer. Thank you to the three of you for joining for our interview today. To get things started, I really wanna talk about technology and human connection. I know we talk a lot about how often AI can get in the way of those human interactions. That's a big concern here. And so, Jennifer, I'm hoping you can kick us off by kind of explaining how you think healthcare organizations can best balance AI-enabled solutions while keeping empathy, compassion, and human understanding at the forefront. Sure. Thank you, Emily. I'd be happy to start off with that question. I think it's important. So we know over the last twenty years or so, we have introduced a lot of technology into healthcare, and we've heard from both patients and providers that sometimes it has been a barrier, in the past and has created perhaps some administrative burden for our teams, that we are always challenged with. I think the hope for AI, in particular, is that we can find opportunities to alleviate a lot of that administrative burden, freeing our teams up to lean in, to be present in the moment with patients and family members, and, to really move forward. And some examples of that would be utilizing AI for communication push, that right now perhaps is a big administrative burden. I think also utilizing AI, which is something we'll talk about today, to better listen broadly to our patients, their family members, as well as our care team, so that we can make sure that we're listening to all of those voices. We're able to really understand our opportunities to improve, as well as what we're doing well, and we will no longer have some of the gaps that we have in that feedback space. I wanna key in real quickly, Jennifer, on the listening piece of this. We know that this is crucial across healthcare, and I'm wondering, what are your thoughts on how we bridge that gap, right? While we include the technology that becomes useful and frees up time, as you've mentioned, but also ensure that we're listening to the unique needs of everyone we're serving. Absolutely. So, I think there's macro-level listening that we know that we have opportunities to do better. We know that we have gaps in, for example, our survey feedback. We can see by age group, by generation, a decline in the responses that we're getting from our survey feedback. So looking for additional ways that we can listen in more modern and contemporary channels, particularly so that we can capture maybe some of our younger voices, 'cause we can see that we don't get as much feedback, for example, from Millennials or our Gen Z folks who are just now aging into healthcare. So that is one component. The other piece is really looking at tools that will enable our frontline care providers to have better, more personalized information at their fingertips while they're actually talking with a patient. So, for example, we have one of those tools in My Story that allows us to bridge that gap between, in that, you know, that short period of time where we have that opportunity to connect with folks, we can capture some information on the front end about who our patients are, what's important to them, what matters most, and make better use of that time when we're in person. It allows our care providers to lean in different ways. Thank you so much for setting us up here. I wanna think a little more broadly here and actually get some insights from the three of you on what's currently shaping the healthcare experience market, and what are some of those unmet or undermet needs? And I'd love for all three of you to answer. And Vinitha, maybe you can kick us off first. Absolutely. Happy to, Emily. So, so many unmet needs in healthcare. We've all been in healthcare and probably deal with it on a daily basis. But if I think about consumers, I think some of the big unmet needs from a consumer perspective is access. Social determinants of health add to that, unmet need even further. The ability to, kind of understand, you know, who to go to when they need care and, you know, once they have that, like, the price transparency around it, like, what's it gonna cost me? On an average, here in the U.S., we're spending close to $13,000 per year per patient. And that's a lot of money, but people have no clue, still as to how they're spending that money, what is it going to cost them. And then once they've had that care, on the other end, billing and issues with billing seems to be a huge unmet need. And then once they are in care as a patient, coordination of care and, you know, managing that seems to be such an under-met need. The whole primary care model helps solve some of that, but still, consumers tend to be the biggest advocates, and they have to ensure that, their information is being shared across their care team in a way where they feel like they are heard. And then talking about being heard, that seems to be a big unmet need. You know, patients are constantly talking about the fact that, you know, "They don't know me enough," right? And, and that lack of personalization is definitely a huge unmet need, across the healthcare continuum. And then, finally, if you think about the workforce, we know about the burnout. It's real. You know, the healthcare workforce has that burnout, and a lot of times we hear about the fact that it's not the quantity of time, right? It's the quality. What are they doing with that time they have in front of the patients? Are they able to maximize that? And that is a huge unmet need. And then you add workflow and issues with workflow in the workforce being a huge issue. So lots of unmet needs and lots of opportunities for us to figure out how we solve that. Yeah. Christophe, I'm wondering if you have anything to add to that, that long list of unmet and undermet needs that you're hearing? Yeah, to not be redundant, if I think about the health system on the leadership side, what I'm hearing the most is too much data and not enough time. I call it the needle in the haystack problem. Like, where should we focus? What should we do? How do we know our actions have expected impact? Absolutely. And to that point, the data piece of it, I mean, I feel like this is where NRC Health can come in, and I'm wondering if you can expand a bit on too much data. How is NRC Health addressing that issue specifically? So it's multiple sources. You have obviously data. We collect more and more data at the patient level. You have information coming from all parts of the health system, but usually those data sources are all disconnected, disjointed. So it takes a lot of efforts, not only to trend a single metric, but when you have hundreds of thousands of them are distributed, putting them together and having insights to understand, like, what are the correlation, causality, it takes a lot of efforts today. So this is one of the key problem we're trying to solve today. ... Absolutely. Vinitha, anything to add to your list on how NRC is addressing some of those needs? The list is long, as we know. Absolutely. So, you know, as a person who builds products for a living, the way I think about, you know, needs and how you solve for it is, obviously, first is acknowledging these needs, right? So these needs exist, and they are unmet, and we are aware of it. We have tried to solve it, we have others in the industry who are trying to solve it, for many, many years now. So the second part of, you know, when I look at problems, what I do is like: Can you solve it, and can you solve it well? And that well is the key point here, is, you know, we can solve for some of these issues, and what we are doing at NRC is really focusing on solving it well. And we'll talk about some of the ways we are solving for it. And then the last part of when I build products is like: Can you build durable advantage? And that comes in the form of scalability and sustainability, right? And so a lot of what we are doing with these needs is really focusing on ensuring that we understand them end to end, and we are focused on solving them well. Building Huey, our AI engine, was part of that. We needed something to pull all those data silos that Christophe talked about, and making sense out of it. And Huey does that, among many other things. And then building that durable advantage of, you know, ensuring we can sustain and scale this across. So not just looking at these problems in isolation, but looking at it as a continuum, right? So how does problems that exist from a market perspective lead to problems that exist at an individual consumer level? And then how does that translate to patients, and then how does that translate to the workforce? So looking at it in that continuum is also something about solving it well, and Huey helps us do that by really helping us connect and look at it end to end. And I look forward to telling you all more about it, but that's really how what we are doing is focusing on solving it right and ensuring that we're solving it not just for now, but for a longer term. Thank you so much. And that excitement around Huey, I can hear it in your voice a little bit, and it sets us up perfectly to where I want to go next here. I'm hoping we can talk about Huey, the why. Why Huey, and how is Huey different from other AI engines? And Christophe and Vinitha, I would extend this question to both of you. Yeah. So, like, you know, like, there's a lot of hype about AI. I mean, like, you can turn the news, it's there every day. And I think there's actually not as much AI in the real world, like, being used as people may think. There's a lot of ideas, there's presentations, you may have a few prototypes. There's a lot of reasons for that. Technology is still pretty new, it's changing fast. Some people have over expectations. I think a lot of people have underappreciations of what's possible. There's also concerns about what AI can do. So the way we're approaching it and why it makes Huey pretty different is it's a technology we've designed purely for the sole purpose of experience management. It's homegrown, and we optimize it for a single purpose, is experience. It's all we're working on at the moment, and we looked at the abilities that Vinitha described, some of the problems that people are facing. When you look and mention, there's a lot of data. You only have two types of data. You have what I call the structured data, like surveys, graphs, metrics, numbers, and you have unstructured data: comments, videos, voice. So we've built an engine that can suck in all this information, have multiple data sets overlaid, and create personal insights that look at the different aspects. So what can you do with this type of technology? You can synthesize large amount of information, you can simplify repetitive tasks, you can detect anomalies, you can maybe help with decisions, you can enhance the learning. All those, the paths that I'm describing are very specific to our industry. So general engines cannot do specific, solve specific use cases if it's not focused on the domain of experience management. And the way we tune the scope of what we work on is really by working very closely with our partners. Actually, I was looking at the attendee list and people saying hello. I see a few names that I've talked to in the last few months. So the calibration of the engine is not coming out of a box randomly. It's through those conversations we have with our partners to really identify what problems do you have, and how do we bring AI in your workflow to minimize the friction, and so the adoption and the usefulness can be very natural. Absolutely. Vinitha, anything you want to add there? Yeah, no, Christophe described it really, really well in terms of what differentiates Huey from the rest of the market. And as we built Huey, our goal really was, you know, not just building this engine, you know, that can allow us to make sense out of all this amazing data that exists out there, but we also took a step back and said: "Are we collecting the right types of data, right? Are we collecting information? Are we listening across the journey? Are there touch points within the journey where, you know, we might not have enough information? Are we not getting enough information from the youngest of our patients, and how does that translate into the impact that we want to generate from it?" So, a big part of what we also did was to ensure that we created new ways of listening, and added the power of Huey on top of that. You know, in terms of the use cases, like Christophe said, we really are listening to the market. There are tons of use cases as we described in terms of the unmet needs, but we focused a lot on the ones that will allow our partners, our clients to try it out, implement it at their system, and start seeing the impact, and then we want to build on it, right? A lot of the use cases are things that allows for some simple problems to be solved and extreme impact and immediate results from it, and then use that to build the use cases that come after that. I'm really excited with what we have been doing here. ... Thank you so much. And, Jennifer, I would love to hear from your perspective, kind of the importance behind the co-designing piece of this, right? When we talk about building an AI exclusively for this use case, healthcare experience management, where does that co-design piece fit in with our partners? Oh, I think it is so important, and it's actually something that even as I am a former customer of NRC, I always appreciated the spirit of co-design. We know that it is by listening and truly leaning in, getting your thoughts, understanding your problems that you're trying to solve, that we can be the best partner in that. And so when we bring, you know, our team together with your teams, we're gonna come out of that with a better product, a product that's designed specifically for our customers and the problems we know that you're all trying to solve. Absolutely. The next place I wanna go here is talking a little bit about ensuring that Huey is aligned, right, with human-centric and societal values. And I would love to hear from all three of you on this one. Jennifer, maybe you can start. Sure. I think, again, there's a lot of hope for AI, and at the same time, there's a lot of concern that we're rolling it out responsibly, and I think the conversation around that responsibility has to do with: How are we aligning with our societal values and what is most important? It's what I'm excited about with Huey, because again, as Christophe mentioned, it is something that has been specifically designed for this space, co-designed with clients, understanding the unique opportunities that we have in healthcare, and those values that we wanna hold of human understanding, and so I think the hope of AI and these types of platforms is that we will be able to roll it out responsibly and ethically. And again, I really think the power of this lies in freeing up frontline folks to actually spend time in relationship, having conversations, understanding who people are, what matters to the folks that we're caring about. You know, I think we've heard a lot of feedback over the last decade or more that our care teams are frustrated with the fact that they don't have the time they once had to really lean into that relationship building. And so I think when we think about AI and that, the societal piece of it, it's about: How do we apply it in ways that allow us to spend more time being human, and maybe being human beings and not being human doings, which a lot of people feel right now they are? Great. Vinitha, Christophe, anything to add there? Yeah, I'll add something, and it goes beyond actually NRC. For me, it's about, like, what myself and all my peers, all the technologists in the world are gonna do with it. One of the core belief that I have, and we have at the organization, is we look at AI as an engine to augment human abilities, not to replace humans, and I'm really hoping that in every single company, this is where we're gonna put the bar. As much excitement as people should have about it, the fear is very genuine. I mean, like, it easily can get a slippery slope where AI will be going places where it shouldn't, and in my opinion, in our shared opinion, the best. People I know that the best are people we work with, and so that engagement is just not—it's just not to say we work with customers. It's really to find where is too much. So this is something we take as a core ground to everything we do. We pass through that human aspect through all the cycles, and that's why, like, also going pretty fast and showing not PowerPoints or ideas, but actually showing products. This is how you get feedback. I personally don't like product roadmaps that, like, put big ideas on paper. I like to show software, because that's how you get the actual feedback from people that use it. It's like: "Yes, this is helpful for me. No, this is infringing on my human rights." All right? And that boundary is gonna come further and further in not only in this current environment, in all our personal lives and work lives, through all the technology that is coming in the next few years. So our bar is very clear. We want to augment human, not to replace them, and the test is with the market. Yeah, and adding to what Christophe and Jennifer just said, right, in terms of actuality and what we are doing at NRC with, you know, those principles, is that, you know, we test our solutions in, like, a two-step process, right? So obviously, we're checking for, you know, values and biases and hallucinations and whatnot in the curation of the data. By ensuring we have the adequate guardrails around our data, we make sure our LLM checks for it as well. And then we follow through all the standards. In fact, we don't just follow through standards, we are in the forefront of it. We are participating with all the standards in ensuring that within healthcare, we are following the goal of having responsible, ethical AI. So that's step one. And then step two, even within our applications, we build our solutions with our partners, like Christophe said, but we also make sure that within the solution, we have all the sources listed as to how we arrived at how Huey came up with an answer, and then we want our clients and our users to actually validate it. Validation is key. And that second part ensures that what we are doing in our first part is done right, and then Huey is constantly learning. So while we do this two-step process, it's constantly evolving to ensure that we keep to that highest level of responsibility when it comes to societal values and ethics. Thanks for wrapping that one up for us, Vinitha. We do have a question that just came in. Christophe, I'll hand this one to you. How are you addressing patient privacy when applying the listening and personalized solutions? And I know we wanted to talk about security and privacy, so let's start there. Yeah. So, I'm not gonna go into architecture detail, but, we have a privacy, what I call a privacy first architecture. So it's built in all the layers. We use a set of open source and homegrown technology. And we have the benefits of running this fully in our own cloud, which is not unique, but fairly unique still. So, first, like, on the outputs, we don't push any data to any third party. Like, all the information coming to NRC stays within our system. So no sharing. We also have a number of layers that take care of the isolation of data, like by customer, by patients, and all of this. And while claims are really good, and they can claim as much as they want about the protection of data, a much more interesting in the next step, Vinitha talked a little bit about it, it's like we will prove everything in terms of security by being independently verified. We've been one of the very first partner to talk to HITRUST on their new AI framework. We've been in beta with their new compliance engine or framework in a while. So we're gonna be publishing, and we're very welcome to be very transparent in how does it work, and have a third party to come and validate that our statements are not again a PowerPoint, but a proven choice. And we follow the White House executive orders and very close also to the NIST emerging standards. So something I'm happy also separately, if anyone wants to have more understanding in how we approach this, but it's the foundation, and the first foundation is augmentation of humans. The second one is to do things right. So we have a great foundation for this. And the last thing I'll mention also, which is not often mentioned, is we have a strong belief in not doing global model training on client data without consent. It's not the case, especially when you use things like OpenAI or other systems. Everything you query is also making their engine better in ways that may not be fair. So we have a policy to not do training outside of the scope of what is being declared. Thanks for outlining those differentiators when it comes to security and privacy. We have another question that just came in wanting to know about the main differences or limitations of Huey compared to ChatGPT. Mm-hmm. Yeah. So who wants to take that one? I'll take this one, so ChatGPT is trained from the knowledge of the internet, and we use similar engines, open source, as a foundation to understand the human language, or I'd say human knowledge in general. We have not shared with our customers yet, but a version that can understand audio and video and all of this, so you understand the content, so you use those general, generic LLMs at first, but on top of that, if you want to understand customer data specifically, you need a custom solution that learns all the data about a specific client and makes sense of this, so you can't, you can't do that the same way. Like, you, you can't push the same type of information to a third party and expecting the same type of answers. Like, our engine is designed and tuned specifically to take the vast amount of data from a, a single customer, and it's been tuned for quite a while now to really understand the domain of experience management like nothing else. Thank you. I would love, Vinitha, if you could talk us through a little bit about what Huey can do today, and then we have a few questions wanting to know where it's actually being applied. So maybe you have some scenarios from some of those early adopters you could share as well. Absolutely. So we started off Huey with doing something very simple, which is allowing it to ask questions, right? So, so you can ask Huey questions about your data based on all the free-flowing comments that we collect today, and Huey is able to create a narrative that makes sense, right? So it's not, you know... It's taking what could take hours for you to scan through all the data and look for trends, Huey can summarize that for you in a very easy, simple way. And so you could ask questions like, you know, "What are my top issues in this particular floor?" Or, "Who are my top performers?" Or, you know, "What should I be focusing on when it comes to efficiencies?" Right? So the questions could be anything. Anything that's top of mind for our clients can be asked, and they can get results from it. So that's the easiest use case. We wanted to start off there, and then what we've also done is we've also made our service recovery smarter, so the ability of not just you know, responding to these comments, but responding to them in a way where things are already created for you. All you have to do, you have all the context available to you, so then you have the ability to actually look at it, edit it, and send it, or have the ability to use it as a training material for somebody who's going to make that phone call to address that issue or the complaint, right, so that was the second use case we launched. We also launched something called nGage, which is our a listening tool. So it has the ability to collect more of that data I was talking about, right? We want to have the ability to ensure that you're listening in areas where you didn't listen before, or you didn't listen enough, like maybe in your cafeteria, or maybe in your you know, you need to know more about your environmental services or issues with parking. So, having that ability to listen more holistically was a third use case. And then having Huey create you know, insights from it, right? So not just listening to it and understanding it, in its raw form, but having the ability to augment with it, you know, context, past history, or just listening to the voice and figuring out, you know, what else, what is the hidden message, or what, what else are these patients telling you about? So that was the third use case. The fourth one, we kind of focused more on, our clinicians, clinician staff, and what we did is they get a lot of feedback, and they have to parse through that on a daily basis. We created another like a narrative out of it, so they can focus on what's really, really critical to them. And, and then it includes coaching tips and whatnot. Again, from Huey, based on, you know, not just, you know, kind of the instance of their information, but Huey building upon it based on, you know, what it has seen, to function and work better across the system. So these are the initial use cases that we have launched, but the ones we are really focused on are the ones that are more personalized, right? So our goal was to start off here, but the next set of applications that we are launching with Huey, that the ones we are working on right now, include things where we have a more personalized view of the patient, ahead of their visit. We already have a solution for that Jennifer spoke about earlier, called My Story, but then taking it to that next stage of not just creating an insight about the patient who's going to walk in, but providing information so that the focus can be on the patient and providing them the best possible care and providing them the best experience possible, and then more personalization in terms of looking at information across the continuum, right? So the data silos are real, so experience management is caught in silos as well. So across patient experience and employee experience and consumer experience, like looking at it across the board and trying to create personalized use cases that allow you to create campaigns or create a strategy is kind of the next set of use cases we are working on. Thank you for that overview, Vinitha. And believe it or not, we've already reached our time here together today. We probably could have hosted this webcast for multiple hours here. I know that we didn't get to all of the questions, so I just want to let you all know we will be doing follow-up to those questions we didn't get to today. All of our panelists are available, too. If you want to reach out to them and continue these conversations, we encourage you to do so. Again, I would point you to those related resources that we talked about at the beginning. Lots more information there. And then you will be getting a recording of today's discussion within two business days. We thank you so much for joining us today, and thank you-