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

 Thanks for joining us on this week in Health It. This is a solution showcase. My name is Bill Russell. . Former Healthcare CIO for 16 hospital system and the creator of this week in Health IT a channel dedicated to keeping health IT staff current and engaged. Alright, today we have a solution showcase and we're gonna talk about some really cool advanced technologies in healthcare that is transforming healthcare.

We have Chris Goff with Intel, Dr. Cornes with Worldwide Technology, and Dr. Eric Quin with Worldwide Technology as well. I wanna thank WW t and Intel for sponsoring this solution showcase. It's a great conversation, uh, a lot of really cool solutions they've come up with for healthcare. If you want to learn more about this, you can hit the website wwt.com.

They have a lot of really great material for healthcare professionals out there. If you wanna be a part of our mission, you can become a show sponsor as well. The first step. It's to send an email to partner at this week in health it.com. Your response to clip notes has been incredible. And why wouldn't it be you helped create it?

Clip notes is an email we send out 24 hours after each episode airs, and it has a summary of what we talked about. It has bullet points of the key notes in the show and it has one to four video clips, so you can just click on those video clips and watch different segments that our team pulls out that we think, uh, really captures the essence of the conversation.

It's, uh, simple to sign up. You just go to this week, health.com, click on subscribe, put your information in there, and you'll start receiving. . Clip notes after our next episode airs. It's a great way for you to stay current. It's a great way for your team to get to stay current and a great really foundation for you and your team to have conversations.

So go ahead and get signed up, get your team signed up, and, uh, begin getting clip notes after the next episode. Today we have a solution showcase. We have Chris Goff with Intel, Dr. Sanaz Cordes with WW T and Dr. Eric Quin with Worldwide Technology as well. Good afternoon everyone, and and welcome to the show.

Thank. Yeah, I'm looking forward to this conversation. There's an awful lot of talk right now around the locus of care and the shift during the pandemic to remote care venues and really into the home, and we're seeing some of the payment models start to follow that. So let's start with this question. I.

What kind of challenges come along with that change? And what kind of challenges does this represent to healthcare and to health IT organizations? So we'll start with you, Dr. Cordes. What are your thoughts about the change in the locus of care? You know, I think that there's sort of the, the tactical logistic challenges, but there's also the mindset, the philosophy of really shifting the way health systems think about care as

Typically it's been episodic and reactive care, right? Patients come in and and access care at points in time, and then we react to it to this model of continuous proactive care. I think that's what the pandemic has shown us. It's really important to be able to kind of make that shift and doing it remotely and being at home is truly the best way to capture that sort of continuous data and find out information before.

Before it becomes a challenge and you need to intervene medically. And then I think from sort of just day to day. Execution of that, the access to the care, beginning with just the digital engagement. Do they have a portal set up properly? Can patients have that retail, the sort of experience that they expect to be able to get on schedule, launch their, their remote engagement, have their devices at home, feed in seamlessly, and then the connectivity.

We serve a lot of health systems at Worldwide that have rural locations or various. Communities that don't necessarily have access to wifi or devices that, uh, can allow them to have some of this care. So making sure that they can do what they can to, to solve for that. And sometimes that involves things like putting up stations in communities, libraries, and things like that, which we're helping folks do to be able to come in and access telemedicine.

And then, of course, for the CISOs of the CIOs and all the folks on the IT side of the organization. The data, getting the data, aggregating the data, putting the data in the right place, and the security around doing that. As more and more devices become part of the remote plan, making sure that that, that patient information is passed through securely.

So, so as a follow on to that is, do you think this is going a permanent shift that, I mean, really accelerated.

As the pandemic has sort of, uh, accelerated it, do you think it's gonna continue at, at this kind of pace following the pandemic? I do think so. I think we're sort of in a different phase now where that kind of rush to have a lot of telemedicine visits and completely not coming to the hospital. As we all know, that's kind of, we're past that and people are feeling safe and coming back in.

And I'm excited to see what that might look like. Yeah, me too.

So the immediate goals of that was to really allow what experienced hospitals to quickly expand care to their Medicare beneficiaries, um, as well as the hospitals themselves are, are required to submit this monitor data. On a per month basis. So that's one, one prong, or one thing I see. Another thing is at the end of February, I believe Humana Med Humana's Medicare Advantage population, we'll have in network access to Mercy Virtual, which is the, the hospital without, you know, without beds.

Out in St. Louis, so staff with greater than what 300 clinicians providing a proactive remote patient monitoring for patients at home in the hospital setting as well. They're being monitored again by Mercy Virtual, and this is really a value-based care play with Humana for their patients in, I believe it's Arkansas, Kansas, Missouri, and uh, Oklahoma as well.

And then lastly, I would say on the, one of the non-traditional, so Big tech, you have Amazon Care, but not alone with Intermountain Health and Ascension Health that you know, that are founding members of the coalition and they're really have been driving and lobbying to make, uh, permanent changes. To the home care health reimbursement policies.

So that's very huge. But they also have additional members that's moving health home, uh, initiative. Think, uh, landmark Health, Amwell, uh, signify Health and several others. So there's a lot of. Forces that are driving, you know, these initiatives to be to Moving Health Home. I think today, if you look at just simple doctor patient virtual visits or remote patient monitoring, the home for chronic diseases.

As Dr. Cordes said, that is those, those require kind of discreet measurements. There may be a few times a day as these acute services shift into the home. Over time, we think there's gonna be a need for mu for more streaming telemetry and continuous data. So that will cause some additional challenges and, uh, technical requirements to accommodate those that, that different kind of data acquisition in the home.

And that's the direction I wanna go. It's interesting that you said a hospital without beds, it actually has beds. There are beds, right? It's our, our beds at home. As A-C-I-O-I used to get asked all the time, how many beds does your hospital have? And I guess for mercy, they have to say this many acute, but they also have to say no.

This many people are being monitored from our virtual hospital in their own beds. Really? Right. So along the lines of what Chris was, was talking about, what kind of conditions are we seeing sort of emerge at this point that people are using this kind of realtime data, realtime tracking in delivering that to a care team so that they can provide care, uh, for that population?

Well, I think initially.

There's a , one can try to boil the ocean and try to look at all chronic conditions. That might be a little too much if you focus on what we're seeing or what I've seen is focusing on those chronic care conditions that we, um, spend a lot of money on, or there's a, a lot of opportunity there to be more proactive with that population.

Those are the kinds of conditions that, that I see. This really . Getting in front of, so for example, congestive heart failure, diabetes, hypertension, and things like that. Things that are some of the most common conditions we see that we're spending our most dollars on in healthcare to be proactive with those patients and taking corrective action before they end up getting at a bad place and having to be.

Off to the ed. Those are the kinds of conditions that I've seen. Chris, I wanna come back to you, Microsoft and Intel engaged WWT to develop a hybrid cloud solution to demonstrate the power of the Azure stack Edge pros, edge compute, and AI technology. Tell us a little bit about that project. Yeah, sure.

Well, as we've engaged with healthcare organizations. A common need for like an intelligent edge solution. That's where the patients are, that's where data is, is originally acquired. And there's some applications or use cases that really are more appropriate for, to be deployed in that physical location.

So that could be driven by latency. So for example, if you're looking at a large imaging study and you want to do some image analytics to find a feature in that data, you wanna make a decision. They at the point of don't.

Video is another good example of that. Complex streams and also regulatory perspective. Depending on what country, state, or province you might be in, there's different rules for how the data is regulated and where it can reside, where it can be processed. So for all those reasons like hybrid, we, we feel like a hybrid cloud solution lets the healthcare organization.

Run certain applications or workloads there, you know, at the edge, at the point of care. And it could be like, for example, fall prevention, analyzing video. We see that use case quite a bit. And then for applications that are more suitable for the cloud, like maybe a longitudinal, you know, population-wide study, but more of a health research use case, they can decide to.

Requirements of that particular application or workload. Do I wanna run this on the edge at the point of care? Do I wanna run this, uh, in the cloud? And that, that flexibility is appreciated. Yep. Absolutely. So many health, uh, and, and Dr. Chu, I want you to comment on this, many health organizations are, are stuck on the challenge of compliance, privacy, latency, hands.

Overwhelming them. How, how does a solution like this address these challenges? Sure. Simply put, this helps in identifying the signals from within the noise. Clinicians are slammed and they're just overwhelmed with the amounts of data that is, you know, humanly impossible to manage. It's sheer cognitive overload by numbers.

Tools like this, get the right data about the right patient to the right clinicians at the right time in order to be proactive and to be supportive in their critical clinical workflows. A tool like this can can be configured, you know, for chronic care management populations, post discharge care populations, behavioral health, even rapid response teams that are in the hospital that are responsible for surveillance.

Of acute care populations, the most rapid response teams are activated after the fact, after something bad has happened. So tools like this can definitely help in real time triage patients and, and get them admitted to the right place in the hospital. An example of that would be a patient being admitted, you know, to the floor for observation from the ed.

Many times there's, uh, a changing disposition with a patient like this because they're waiting, you know, for bed for a long period of time. Having a tool like this to be able to bring these insights to the forefront so clinicians can make adjustments and make sure the patient gets transferred to the right level of care, perhaps the ICU.

It is very important, and these are things that we need today to really change the outcomes, but time is everything. Early recognition of patient decompensation driving early intervention leads to better outcomes and cost and, and all this can be done using the organization's clinical protocols and even developing advanced analytics models that are tweaked by the organization's governing bodies.

Talk about how that is even possible. There's nothing we love more than a deadline challenge worldwide. Yeah, no, it was phenomenal to watch the team in action. I mean, we have such an, a diverse team. We're really fortunate to be able to just quickly assemble, um, the right folks. So we were able to bring

Physicians with practice experience like Dr. Um Quinones, myself, physicians and nurses who have health tech experiences, who deeply understand how the electronic workflows work. Where does information need to go? What needs, as Dr. Q is saying, what notifications go where, et cetera. We have data scientists.

We have dozens of PhD data scientists on board. We have AI ML experts, our software developers, we have UX designers, product managers. So we were able to quickly assemble this group, this really great team to start working in a sort of a iterative model. To identify the use cases, vet them, have really great collaborative discussions.

And you know what? Worldwide, our application services team, we have agile methodology to see that in action and see how, um, valuable that is to be able to, to roll things out this quickly. I think this, this project really brought that to light so. Yeah, we did it. We were able to kind of roll that out and we also were able to tap into some of our customers.

We at worldwide touch one in five patients in the us When you look at our reach of, of our health system, um, customers. So I actually was able to meet and I think Dr. Kon, you might've as well, met with a couple of our health systems organizations, CMIOs and others to, to get some feedback along the way, which was really helpful.

Give us an idea what was developed. So clearly you didn't develop it for all chronic conditions. It was probably focused on one. Given that timeframe, what was actually developed and what did it showcase? So we, uh, so first of all, clinically, the use case that we landed on was heart failure, heart failure's, costliest condition in the US to care for, I think, $70 billion a year.

Spent the number one cause of death in the US So we sorted through several chronic diseases and that one made sense clinically. And also as we looked at what's a data rich environment, uh, for someone potentially at home. And of course, this solution can work in a lot of other settings, and we can touch on that, that's not just at home, but in this first model.

A data rich environment, a heart failure patient may be able to, we would be able to monitor multiple devices like blood pressure, cuff and scales, and pulse oximeters and EKGs. So it made sense that this would be a good starting point use case. And so starting with that, we were able to create protocols and Dr.

Quinones led most of that effort using . Practice guidelines and things through evidence-based medicine to sort of create the protocols, um, of what it would look like as we monitor a patient at home, the normal abnormal. And what's the beauty of this solution? Abnormally trending normal, which you don't normally get to do when you're doing point in time.

Data collection. So when a patient comes in, you're just seeing their blood pressure when they're there and it's probably high because they're dealing with white coats and the stress of being at a doctor's office. And so creating, you know, the solution so that the AI and machine learning could then come in and say, well, when the pulse ox is still normal, but it's trending this much below where it was the last five days, and the scale indicates a slight weight gain, that's still normal.

And you marry that with X, Y, and Z. Now we are getting some data rich insights and now we can take proactive, um, steps. And so that was all kind of put in place. And then from the actual user interface perspective, we had a very beautiful design that our app services team created. That was sort of the red, yellow, green model of showing trends and then being able to actually show notifications.

And it was a, a, a great dashboard that was configurable to, to what physicians want would want to use, or case managers or anybody that would be accessing the information. And that was the initial sort of prototype that we rolled out for the Ignite Conference. Fantastic. What potential outcomes for care, as Dr.

Cordes was talking, you can apply this at the home, but you could also apply it within other care settings. What kind of outcomes for care do you anticipate as a result of this type of solution? I, I think taking a step back a little bit, thinking about where it actually helps, right? So if you think of a solution like this really helps us scale and it, it helps us to have greater optics for the patient and the population's longitudinal journey.

Clinical resources are thin right now. We see the numbers that are entering med school that are starting to increase a little bit, but.

So we need advanced technology solutions like this to help bridge that gap. And, uh, integrating, you know, multiple data points, not just the data points that talked. Data points become greater. Not just by those labs, but very wearables, the genome, social determinants of health, environmental and ecological data.

This all drives us down that proverbial precision medicine road that we're all striving for. And I think this can have a tremendous impact on cost, um, outcomes, patient satisfaction, and clinician experience for sure with this level of data can also help drive. The acceleration of research and life science.

So there's a lot of outcomes that I think that are gonna be coming from this. This is a great example of a, a partnership between Intel and WW wt. How, how can people really tap into the solution? If people are interested in this, how can they find out more information? Yeah, they can absolutely find out more.

They can reach out to anybody on this . On this podcast right now, Chris, Eric, myself, we also are in the Microsoft marketplace, so there's a engagement there that folks can review and, and click on and, and sign up for it, where we can help 'em with learning more about the solution and customizing it and ultimately potentially rolling it out for their use cases.

I.

The podcast, but it's really interesting and I think it also accentuates your, your partnership and that's the machine learning in radiation oncology at at WashU School of Medicine. Who wants to give us some background on, on that project? It looks really fascinating. Worldwide technology. We, we collaborated with Washington University School of Medicine in, in St.

Louis. Uh, to utilize machine learning to really develop methods to predict radiation treatments for patients with, you know, cancer. About one half of all cancer patients receive radiation therapy as part of their treatment plan, and radiation oncologists coordinate with oncology specialists to review appropriate treatments for their patients, what we call multidisciplinary tumor boards.

They use national and international guidelines, patient history details, and of that patient history and, and physician clinical experience with various treatments and, uh, available resources that they have at at their facilities. The radiation or radiology, oncology treatments have become much more complex as a result of all this precision of this type of, you know, treatment is really critical and needs to be unique to the patient.

Radiation not only kills cancer, but it also kills healthy tissue as well. So there's, you know, no room for delays, you know, in treatment time is critical. So this led to Washington University and worldwide technology working together using high fidelity diagnostics, clinical and treatment data to author machine learning algorithms that quickly predict the optimal.

Radiation oncology treatment orders for head, neck and, and lung and prostate cancer using this platform, radiation oncology, oncologists, specialists, the, that, uh, or specify, excuse me, the, the details of the cancer. And they, they specify the details of the cancer of the patient and then as a result, the machine learning algorithm is.

Given the radiotherapy planning target volume, the frequency of treatment, the prediction percent with explanation, which is very important for the clinician, the patient body positioning during treatment sessions, all these things are super important and the type and frequency of radiology imaging needed.

So, uh, that all takes a lot of time and, uh. So this really has a potential to expedite a time consuming process in radiology, oncology orders with the goal of greater efficiency as well. Also from, I, I believe Washington U School of Medicine is authoring an academic paper on this. So we're really excited to see what comes of this amazing collaboration and technology.

Chris, as I listen to this, the, the precision aspect of this, getting to the end of one, as they say. This, this kind of precision, these kind of things create massive workloads, and they're compute intensive. They're storage intensive. I mean, the processing obviously is significant. What advances has Intel made to handle these kinds of workloads and what can we expect moving forward in this area?

This is exciting stuff to get closer and closer to that promise of precision medicine. Absolutely. We work very closely with the, the healthcare ecosystem, both the partners that bring solutions to market, but also with what we call the end users of our technology, hospital systems, pharmaceutical companies, et cetera, so we can understand their requirements.

We, in this case, in the case of what we're talking about today, understand how they use or could use machine learning to improve what they're doing, and we'll, we'll weave those requirements. Right into our technology that we're bringing to market. So we're known as ACPU company. We actually have an XPU strategy.

So depending on what kind of artificial intelligence it is or what kind of machine learning it is, there could be different hardware that's more appropriate for that task. And then beyond the hardware, working with the software ecosystem to optimize all that software to make sure it's scalable and efficient.

Working very closely with partners like WW t and, and Microsoft to weave that innovation into the solution, like the solution we talked about today. So that from the end user's perspective, it's, it's sort of invisible. It just works high performance and, um, it can accommodate the machine learning task.

Whatever task, uh, is, is put to it. So, yeah.

Artificial intelligence. I wander back to some of the conversations I I've had with physicians and there's still a reticence by some physicians to adopt this kind of stuff. How, how do physicians view this and the use of this, of this technology and what's gonna take it forward, do you think? That's interesting.

I. Started my health tech career in clinical decision support. When I stopped practicing and I moved over to the corporate space. I worked for an organization where we sold just static order sets, just CPOE order sets in the early two thousands, and that started my journey of . Adoption, and it's really interesting to look back now, 15 years later, 16 years later, and we're still.

Doing that, right? I mean, if you asked a physician now like, how do you feel about just a passive asthma order set or cp? They'd say, yeah, of course I've been using that for over a decade. But at the time it was very challenging, right? To get this adopted and become part of the workflow. And so I think we're just same story.

Fifth verse. Now as we see more and more AI and ML and that becomes more commonplace and more organizations are bringing it on board so there's nothing magical, it's just the same kind of formula. We need to make sure that whatever solutions we bring forth, in this case, AI ml, there is a, that we're able to demonstrate a solid clinical ROI 'cause that matters.

More than anything to the physician. So having those success metrics in place. So with the case study that like Dr. Quinones just described, with radios being able to point back and say, with a 97% accuracy, this, this, and this was identified in these 200 patients and by measurement their length of stay or any other adverse outcome.

Was decreased by X percent. And so I think that we know that now, right? Like two, two decades into it. You really need to present that to clinicians. And so I think health systems should keep that in mind as they roll out solutions, putting in those metrics and a plan to measure them. And then I think engaging stakeholders early and often, bringing positions to the table, making sure that their voice is heard and that they can help.

Become part of the process and become sort of invested in it, I think is really important. So it's nothing magical. I think it's just what we've been doing over the years. And I guess the third thing that helps is there are the folks that are early adopters, they embrace change. They want these solutions, and so getting them to be advocates to their peers has always been really helpful to me.

Bringing solutions to, to organizations. And so, yeah, I think a combination of those things and we just keep learning from what we've done in health tech over the years and growing on it. We found that when we are rolling out new technology, like say like a machine learning model that's gonna predict, uh, patient risk for some, uh, type of condition, if it can be done in such a way that it minimizes the disruption to the current clinical workflow, like maybe putting that risk score right in the electronic health record as opposed to requiring.

The clinician to interact with some second application that also can help with, with adoption and receptiveness quite a bit. Chris, that is such an important point. Absolutely. Physicians, nurses, clinicians, spend, I. 10, 12 hours a day in the EHR, that is their user interface. The thing I've learned over, over the decades is, yeah, they absolutely don't wanna open seven other apps.

They do not wanna have multiple user interfaces, and so being workflow congruent is absolutely crucial. So great, great catch on that. We've talked about how people can get more information on these solutions. But as a former CIO, I'm sitting here thinking, what if I had an idea? What if our health system had an idea?

Is this the kind of thing that we could tap into you guys and say, look, we've got this problem, we've got this challenge. We're thinking of this. Could we give you the same kind of challenge saying, Hey, it's January and. By, by March, we need to roll out a solution around this. Absolutely. We have 500 engineers.

We have several dozen PhD data scientists, and when they're not involved on an active engagement, which doesn't happen often, but when it does, we have a very robust r and d mentality at worldwide. And so we have actually done recently a lot of work around computer vision, reverse image lookup. In the MRI brain tumor space.

And that was all based on kind of what you just described, bill, where we had a health system express interest in that and ask questions about it. And we brought that back to the organization and those guys just took it and ran and the AI ML team, our business analytics team, and we're able to, you know, build the solution.

And so we are always excited to discuss solutions and, and work with our, our customers. And again. We can reach out to Eric or myself or anyone at Worldwide in, in our organization to learn more about that. And one more point to Dr. Gore's comments. So, at Worldwide Technology, we have a platform that really helps expedite this stuff.

So the, the Advanced technology platform, it's, boy, I think the initial investment was about a half a billion dollars with our OEM partners, such as Intel and others, you know. They have the ultimate sandbox, if you will, and so a health system can duplicate their environment within the ATC, and if they're planning to bring on new technologies, they can do that or they're going to do data migrations or any core IT infrastructure changes and things like that.

They can do it quickly. AG with an agile mentality to be able to come out with an outcome. And help them with the decision. So that just seems, saves time and money. Fantastic. And they can get more information@wwt.com. Yes. I, I'm excited about this. I'm, I'm glad guys came on to talk about this. I mean, these two solutions are interesting in and of themselves, but I love the fact.

That we can brainstorm together, that we can get into a room with, with data scientists, engineers, just different partners, Intel and others, and come up with solutions. And that's what it's gonna take. It's gonna take that creativity of the group getting together. And bringing our skills together to, to come up with new ways.

I think the pandemic taught us that we can do things that we didn't think were possible before. I hope that pace of innovation continues. So thank you very much for coming on the show. I really appreciate your time. What a great discussion. If you know of someone that might benefit from our channel, from these kinds of discussions.

Please forward them a note. Perhaps your team, your staff. I know if I were ACIO today, I would have every one of my team members listening to this show. It's it's conference level value every week. They can subscribe on our website this week, health.com, or they can go wherever you listen to podcasts.

Apple, Google. Overcast, which is what I use, uh, Spotify, Stitcher, you name it. We're out there. They can find us. Go ahead, subscribe today. Send a note to someone and have them subscribe as well. We want to thank our channel sponsors who are investing in our mission to develop the next generation of health IT leaders.

Those are VMware, Hillrom, Starbridge advisors, Aruba and McAfee. Thanks for listening. That's all for now.