Hi, everyone. This is Lucas Voss with Becker's Healthcare. Thanks so much for tuning in to the Becker's Healthcare podcast series. Today, we're talking about enterprise imaging in the cloud, and I have the perfect guest for today's episode to discuss this topic. Eitan Eshner, regional category leader of radiology informatics and integrated diagnostics at Philips North America. Eitan, great to have you. Hey, Lucas. Thank you so much. It's great to be here. Thanks for the invite.
Absolutely. Excited to talk about this. Very, very topical and very on time right now. I wanna start off with introductions here really quickly. If you just wanna tell us a little bit about yourself and and your work in health care. Yeah. Absolutely. So, you know, I have been working in health care for getting close to twenty years now. I think I'm at 18.
I spent the first year and change of my time working on EMR deployments, primarily meditech, but the remainder of that time about sixteen ish years at this point in the medical imaging space, working primarily on packs and VNA and generally enterprise imaging system deployments. And that experience has very fortunately given me the opportunity to travel all over the globe. I spent three years working in China, on imaging deployments. I spent seven years working in Singapore on the same.
And now I've been back in The United States since 02/2017, working with Philips here to, you know, continue the enterprise imaging journey in the cloud and otherwise in North America. Well, it's great to have you. Thanks for bringing your almost two decades of experience to the podcast. It's great to have you. And I know there's a lot of conversations around the cloud, so to speak, or the topic of cloud. Right?
What are the three primary reasons right now customers should consider moving to the cloud and even think about it? Yeah. I I think there's a a number of different reasons, and a lot of them align to what is the actual value of the cloud in the first place. But but first and foremost, I think it it's appropriate and notable that pretty much every vendor in the imaging space is shifting to the cloud as a primary focal point of their ongoing strategy.
So I think in some ways, although that maybe is not a good enough reason in of itself, it sort of necessitates the evaluation by customers as to whether or not from an enterprise imaging space anyway, they will, take the leap and move into the cloud since that is now where most of the major players within the industry are focusing their investment energy and see the future of health care really playing out in the imaging space and otherwise. So so first of
all, that's just an industry direction. Right? But be beyond that, you know, the cloud clearly provides certain very tangible value propositions, and I think they're pretty well known at this point, but I'll talk through them anyway. So first of all, durability and scalability are one of the major advantages of the cloud. Of course, that needs to be considered in
the context of any application that's moving. But generally speaking, the cloud offers technologies to really scale out imaging platforms or other health care, solutions, and also provides a base platform of technology that enables, you know, a more durable and available solution that, is likely to better serve the customer, in regards to ongoing clinical operations.
So so that that's one really important key component, just the durability and scalability that the cloud offers as a foundational technology layer. The next bit is security.
Unfortunately, I think we're all relatively familiar with the increasing frequency and, also damaging nature of cybersecurity attacks, as well as the, I guess, increased sophistication of those same attacks and that health care institutions, providers themselves have actually been considered a prime target because of the urgency of restoring up your operations, the, value of the data, the penalties associated with
losing control of that data. And we have, obviously, as an industry, come a long way, and there are many customers that do an excellent job within their own IT infrastructure and landscape, securing the applications and their data. And, of course, it's top of mind now. But even for those most gifted, most invested, most forward thinking customers, providing the same security controls and technology at the same scale and cost and efficacy that the hypervisors are able to provide it
is no small thing. It was fundamentally impossible for most customers to get there. Yeah. Yeah. And that that's not to say that just by moving your systems to the cloud, you get security. You know, an irresponsible cloud deployment is just as bad as an on premise deployment that is not secured. Nonetheless, the tools are there, and they're affordable and sort of democratized for a wide range of customers and competencies to adopt.
And and so it is just, I think, generally a better place to secure your data, and and, you know, run your IT stack as it pertains to the security question. The final one, which is probably the most interesting one, is the opportunity to innovate and deliver better clinical care in the long run. And I I think that's for many health care applications, including in the imaging space, sort of
at the first stages now. Right? It's not again that you move to the cloud and all of a sudden you've innovated and you've gotten all these new clinical care functionalities and you've really changed the way you deliver health care immediately.
But there are obviously emergent technologies and opportunities also for sharing data, you know, across borders between customer sites, you know, even for a patient with their their families and other physicians that the cloud offers that probably over a long enough time horizon will truly impact the way that care is delivered. And translating all of those technologies back to on premise deployments, while it may not be impossible, like, you know, the obvious one here is AI, it's not impossible.
I think most of us know at this point that artificial intelligence sort of lives in the cloud. Right? You're talking about huge infrastructure and compute requirements. You're talking about potentially hundreds of different applications that have, you know, their own foundational models associated with them and that delivering that sort of infrastructure and technology locally on premise for, again, most
IT organizations is not possible. And even for the ones where it might be possible is certainly not desirable just because of what it means from an overhead perspective for hardware and management and licensing
and so on and so forth. So I I think those are the reasons that, again, durability, scalability, security, and the the opportunity for clinical innovation and to change the way we deliver health care in The United States anyway, really globally, are the the main considerations that customers should be looking at. And then I love that you prefaced this by saying, you know, this is seems pretty obvious. Right? We've talked about the cloud for
for quite some time. People should know some of these factors that you've mentioned because they are so important, like cybersecurity, like the clinical implications of it, which are so key. But the fact is that a lot of folks don't. Right? A lot of folks are not where they need to be from from a knowledge perspective and and really knowing how to operationalize some of this technology.
What are some of the ways that health care providers can best leverage some of this cloud store data that that's there to really then enhance specifically enterprise imaging applications? Yeah. You know, there's, again, a a number of ways that you can approach sort of the opportunity with the data and what the cloud can eventually offer beyond their building scalability, but now getting into more of the innovation and insights perspective.
And, you know, I'll I'll say to start off with that, we've entered over, really, just the last two years, very interesting period where, you know, I I think we've been going through, especially in enterprise imaging, and I guess you could say also in the EMR space, What I would just call is primarily digitization. Right? The EMRs came about, hacks came about, other imaging disciplines followed, but we've we've mostly digitized data up to this point.
And a lot of the reason for that was this idea that we would unlock the data, that we would drive insights, and that we would really, again, enable a better understanding of how we can treat complex cases and improve health care and reduce costs and all of these things that was the promise of digitization, that actually turned out to be rather difficult
to approach. Right? I mean, we we not to say there are no insights as a result of EMRs and and packs and all these things, but I don't think anybody got as far as we would have liked
it to. But just now, in the last two years, with really generative AI coming onto the scene, you now have this opportunity for the first time in history, I would say, to really put, like, intelligence or reasoning to disk and do more with the data than we've ever been able to do through standard data analytics opportunities. And so, you know, that is, I would say, the sort of, like, golden opportunity that exists out there now is how can I take all of my data?
And it's not really just about imaging or about just radiology. It might be cross functional across multiple ologies in the imaging space and also probably combines medications data and, you know, clinical indications and reasons for orders to be done and basically a wealth of information that can come from any number of different sources, and then have these AI models basically process more data than we mere humans are fundamentally capable of managing and thinking through ourselves.
And and then that really creates an amazing opportunity for, again, that sort of clinical advancement in in the way that we we treat patients and manage costs in The United States, at least, where you can start to drive more automation and better decision making with larger sets of data because of really the capabilities of the combination of the cloud and AI that enable you to understand more about, you know, what is the next best steps, how can I drive
a Gentic follow-up for an imaging case, and so on and so forth? And I think that's really where the combination of the cloud and, you know, large language models, multimodal models will will eventually take us and and where a lot of the investment research focus is across the industry today. Yeah. Absolutely. You mentioned something very important in your in your previous answer here around cyber security. Right?
You can there is promise when it comes to cloud storage, etcetera, in terms of providing more safety, but it has to be done right. It has to be something that is that is thought through that the implication, the application is is thoughtfully done and successfully done. I'm wondering what some of the key challenges you're seeing are that that health care institutions, health care IT teams encounter when they're migrating
enterprise imaging systems to the cloud. Right? What are some of those things that you're seeing that that are challenges? And then also, what's helping address them right now? Yeah. I I think there's, again, a number of different challenges that are being seen, but the the most common ones from an from an IT perspective anyway in terms of adoption are, you know, first of all, resourcing.
We talk a lot, especially medical imaging, about the shortage of diagnosticians, whether it be radiologists or digital pathologists. We talk about technologists and, you know, this really quite incredible gap in availability that there are for actually performing image acquisition within radiology department today across the country. But we don't talk as much about IT organizations and their ability to find
the right skilled resources. And, of course, it depends a bit on geography and also policy in terms of who you're willing to hire and where you allow them to be. But, you know, in from the perspective of a, you know, seasoned health care technology professionals, the cloud is relatively new. And health care IT is not probably well known for being an early adopter of new technologies for for good reasons.
And that may be changing slightly now, but actually recruiting knowledgeable technology resources in all geographies over the country is is not so easy. So that that's one major challenge. I think the good news there is obviously that a lot of the industry, at least in medical imaging, and I think just generally so, is positioning itself to address that challenge in terms of providing, software as a service based solutions that really mask the technology burden for the customer.
So, you know, and for Philips, for example, the opportunity for us to host and manage our own, you know, homegrown and developed solutions really means that the customer doesn't carry that IT burden. And that is, from our perspective, our core competency. Right? We know these solutions better than our customers do. We built them, so that should be the
way. And so we are able to provide them to the customer, you know, in a way that provides them with the best and highest quality service with, you know, frequent updates and all these additional service entitlements that are very costly to achieve
as an independent customer. And and and we're able to do so, hopefully, at a, you know, in a financial model that scales across our wide range of customers that provides a better total cost of ownership to every customer because it's sort of being shared as a service entitlement across the install base. But that that does bring me actually to the second of the major challenges, and I see this all the time in conversations with customers is cost justification and return on investment.
It's hugely challenging in the cloud environment. A lot of customers, again, look at, you know, should I self host my solutions? What does it mean from a compute perspective, from a storage perspective, from a data, movement perspective, ingress and egress?
And, obviously, every solution has different patterns and even, you know, different clinical disciplines have different patterns of access, in terms of how often the data is accessed and is reviewed and how much of us will go in and out of a cloud based solution. And so as a customer evaluating your existing IT landscape, it can be very challenging to really understand what are the metered costs of the cloud solution going
to be. Right? It's very different cloud economics from what most solutions in the health care space have been sold at as in the past, meaning most of the time solutions are sold almost as a break fix SMA. I buy software from you. I provision my own infrastructure. I can write down on a piece of paper all the servers and switches and, you know, cables and storage that are required, and I can roll that into a cost. And then there'll be a maintenance fee, and that's my
cost. It's very easy to put down on paper. The cloud with its sort of dynamic, you pay for what you use metering model is much more complex to estimate to to derive an actual cost of ownership for, which, again, is another benefit of a software as a service model for customers who are willing to entertain that. The the vendors are usually in that model, again, taking over the risk of the operational run cost of the environment. We are making
the assumptions based on our software. And what we know about it in common customer usage patterns and, of course, you know, information that we get from a customer during, you know, discussions prior to proceeding with a project so that we can appropriately estimate those costs and actually take the ownership and risk if we get it wrong. So, anyway, that that's that's another piece of the puzzle, the cost justification and return on investment.
And I should probably say that oftentimes, the discussion leads itself in my experience to, oh, this is a lot more than I'm paying now to Philips or any other vendor. And, of course, that's true. The cloud from a current spend to future spend perspective may look more from a single vendor perspective, but you have to look at the full picture as part of the justification. Are you going to offset data center costs? Will you be able to repurpose resources that
are actually running your data center? Is there a real state investment there? You know, will you reduce your spend on securing your applications on premise? And so creating a ROI or cost justification model that's more holistic across many different solutions and considers all of the actual service entitlements you would get from a vendor if you're going down a SaaS path is is something that I think takes some educating and isn't maybe the most familiar for everyone in the IT space today.
That's where the partnership piece comes in that you mentioned. It's so important to have that partner to be able to make some of these decisions and have those conversations. Hey, Tom. We colored so many different things here today in our conversation. You truly brought the two decades of experience to our podcast, and I wanna just thank you for for your time and insights today. It's great to have you. Thank you so much, Luis. I really appreciate you joining. Had a lot of
fun. Yeah. Absolutely. And we also want to thank our podcast sponsor, Philips. You can tune in to more podcasts from Becker's Healthcare by visiting our podcast page at beckershospitalreview.com.