Revolutionizing Healthcare: Cloud, AI, and the Future of Diagnostics with Simon Omer, Chief SaaS Transformation Officer at AGFA HealthCare

[00:00:01] This podcast is produced by Outcomes Rocket, your healthcare exclusive digital marketing agency. Outcomes Rocket exists to help healthcare organizations like yours to maximize their impact and accelerate growth. Visit outcomesrocket.com or text us at 312-224-9945.

[00:00:29] Welcome to the Chalk Talk Gym Podcast, where we explore insights into healthcare that help uncover new opportunities for growth and success. I'm your host, Jim Jordan.

[00:00:46] Welcome back to Chalk Talk Gym Podcast, where we explore the cutting edge of innovation in healthcare and technology. Today's guest is Simon Omer, and he's a seasoned leader with over 20 years of experience transforming the healthcare industry. From his work at Philips Healthcare to his current role leading SaaS and cloud transformation at AGFA Healthcare, Simon has transformed and been at the forefront of diagnostic imaging, oncology, and informatics.

[00:01:13] His expertise lies in integrating advanced technologies like AI and cloud computing to revolutionizing patient care and healthcare operations. Now, in this episode, we delve into the data-driven solutions that are reshaping healthcare workflows. We talk about the challenges of legacy systems and the future of integrated informatics.

[00:01:33] Simon shares insights into balancing efficiency with innovation, the role of AI in diagnostics, and how cloud adoption is accelerating change in a traditionally slow-moving industry. Whether you're curious about precision medicine or digital pathology, or the role of automation in improving patient outcomes, this is the conversation for you. So, Simon, tell me and the audience a little bit more about yourself.

[00:02:01] Yeah. So, like I said, Simon Omer, originally from Israel, so moved to the U.S. in 2013, 2011. So, I've been here a little bit more than 13 years. I spent almost 20 years at the healthcare industry doing different things. I worked most of my career with Philips Healthcare. And there I spent time in different areas of the healthcare space from diagnostic imaging.

[00:02:25] So, my first set of products was developing CTs and PET CTs and other diagnostic imaging equipment. And then from there, spent some time in the oncology space. And then my most recent stop was within the informatics space. The thing about it is aggregating, going upwards from a data space perspective, from the images all the way up to everything. And where I'm today with the informatics space is everything coming together into one picture. Today, I work for Agfa.

[00:02:53] And at Agfa, I lead our SaaS and cloud transformation. So, Agfa is on a journey, just like many other vendors in our space, to go from legacy on-prem technology and ways of consuming products and services to more subscription-based and in more advanced cloud technologies. And that's where my focus is, and making sure that both Agfa and our customers are successful through that journey.

[00:03:19] Actually, you've worked for two companies that have both a commercial and a healthcare side. How does your organization fit within the continuum of healthcare today, since it has a little bit of both, but even your previous company did too? Yeah, so maybe we'll start with Philips real quick. So, in Philips, even though it always had different heads, there was a real strong focus within the healthcare division to be focused on healthcare.

[00:03:44] So, yeah, there was overlap in between, and there was the ability to leverage some lighting solutions into healthcare, for example. Right? That's pretty cool when you can do that within the same company. But still, within healthcare, you're 100% in healthcare. And over time, as I'm sure others can see, Philips basically became just a healthcare company, and everything else got divested out. Agfa is not in the same place. Agfa has three divisions. Some of them are commercial.

[00:04:08] But our healthcare division is centralized and focused on what we do, and it's like its own business to run in that space. So, my focus is 100% in healthcare and within the healthcare, the imaging and informatics space. But I think one of the things, I worked for J&J that obviously is famous for consumer and having pharma and medical devices and diagnostics and all that.

[00:04:31] And you're explaining here we would have our focus divisions, but we also had insight into some consumer things that maybe a traditional medical device, a pharma-only company wouldn't, which I always thought was very interesting. So, when you talk about Agfa today, how does that fit within a typical healthcare system? What's it servicing and what's it doing? Like I said, if you think about the aggregation of data, then you start with an image or a set of images and then you go up.

[00:04:59] So, we today operate at the topmost layer where everything comes together. So, when I think about a healthcare continuum, it has everything from patient, early detection, whatnot, all the way through treatment and follow-up and so on. We operate in a very wide part of that spectrum because we will see the patient as soon as the patient comes in for the first type of examination they're going to go through.

[00:05:22] And with the way we interface the healthcare systems around us, we actually have a way to show the entire life cycle for that patient through the system. So, when a physician wants to look at the patient over time, they can actually see the evolution of different stages, whether it's different imaging or different other modalities that went into play, integrations with pathology and other things that come into play.

[00:05:48] So, we actually have more or less a continuum and then we will be part of even the follow-up and post-treatment stages when you think about oncology. Those are the patients that typically ends up having a very long cycle or staying within the system for a long period of time. Then we would actually be able to see an oncology patient over time. Our systems are capable of looking at lesion changes over time. So, we do provide a very broad view of the journey for that patient.

[00:06:17] So, we don't focus just on any more acquiring an image set, but we do look at the end-to-end process. So, I think for the audience folks that aren't hospital-based, they probably heard the word PAC, P-A-C. So, that would stand for Picture Archiving and Communication System. And it seems to me that when you think of information technology in hospital systems, that actually might be one of the more mature ones that are around. Are you bigger than a PAC? Am I defining that too narrowly?

[00:06:47] So, PACS is an element of what we offer. In a sense, there are flavors of PACS, right? You can consider just the archive piece, which is also a product. Some companies will just offer a standalone archive solution. Others would offer both archive and viewing solution.

[00:07:07] Where the solutions get broader is when you start integrating more than just the image administration, but you also integrate in their whole process of treating a patient. So, in my old days, when I was doing CTs and whatnot, my world was there is a scanner. There is someone working on that. They're producing a set of images that are done. Perfect. What next?

[00:07:30] Here, the perspective is an image is just one small piece of information that goes in. And then that image can get processed a hundred different ways. It can be integrated with AI. It can be visualized and then put side by side with other supporting information. So, all that comes together into what we sell as part of our enterprise imaging solution.

[00:07:52] So, it's more comprehensive than, I guess, an imaging or an archiving solution because we do have a full end-to-end solution integration with reporting. So, the reality is that when you go to a hospital and you look at the radiologist where they spend most of their time, they spend most of their time in our system. Because that's where everything comes together from the history for the patient, for the imaging. The reporting solutions are going to be there. They are going to do their dictation in there. They're going to sign up their report. They're going to send it out to their referent physician.

[00:08:21] If their referent physician wants to look at the image set, they're going to connect back to our system to actually look at the images. So, it's all there throughout that continuum. What strikes me is that you need data before you can start doing AI and analytics and all that. And one of the exciting things about the PAC system being one of the older systems out there is you have that layer that you can start exploring. And I was working with a group of people at CMU. This is just more of a researchy kind of thing. But what they did is they took the cyber surveillance.

[00:08:51] Okay, we have a crowd of people. The person had, say, glasses on. And then they had a green shirt. And be able to define that. And they put it over some PAC information at UPMC to say, we have a spot on a lung. And it's too small for us to even get a biopsy to determine what it is. And there's no clinical studies that suggest it's anything. But in our system, do we have any historical data that shows that something like that grew?

[00:09:21] And what did it become? And so, these are the queries that can be made. And my sense is that is what you're starting to realize if I hear you. A lot of what we do is bringing these things together. Traditionally, most of the old interpretation would have been done by the radiologists and the supporting staff. And with the tools that we're able to integrate to do, we're able to do exactly what you just said. Hey, this is something that was identified somewhere else. And therefore, you want to probably take another look at that.

[00:09:49] So, we do things from as simple as helping prioritize which cases to look at. Yeah. To making sure that you didn't miss anything. How about this? We offer an integration. We don't have a lot of our own developed AI applications. But we offer an integration with all the industry best available applications. And we have our own marketplace that we can offer them through. So, from a consumption perspective, it's really easy for the user to say, hey, I want to use this and that.

[00:10:19] And the applications are developed and they can do their own testing and whatnot. And then they can start using it clinically. More and more, I expect us to see those type of aids getting integrated into the work a lot more than they are today. And at the end of the day, the goal there is to help produce an improved level of care, right? It's not going to replace the radiologists with what they do, but it can help. But it also helps free time to focus on the cases that really need more attention.

[00:10:48] Because today, when a radiologist looks at a case, they don't know up front, is this going to be a benign case? There's nothing to look at, but I still have to go through a thousand slices to go through that. Or, yeah, I can probably spend two minutes just to confirm a few things and move to the next one. So we, in a sense, make time for the important cases. And that's one of the commodities that's probably radiologists have the list today. And not just them, but in the healthcare space.

[00:11:14] I think that's what's so exciting about what we're moving to, because I think the first slice is, oh, we're going to replace doctors or we're going to do this. Doctors have questions that need answers. And then when they've got those answers, they deploy a judgment and then they get more questions. And that cycle today or 10 years ago, the asking the question and finding the data to support the question was very time consuming.

[00:11:41] And so I think the efficiency comes as being able to serve those up. Could you become a precision medicine model with your data? Is that something that... Good question. Eventually, I think, yes, I'll take you one step back to my history. The earlier days of AI, when there were not many out there, we actually had a team that was developing something just for that in the space of precision medicine.

[00:12:07] So when I was in oncology, what we wanted to do was to find out whether a certain set of treatment profiles is going to result in a better or worse outcome. We looked at toxicity as one of the main outcomes. They're right. You're treating the right things, but what else are you hurting along the way? So that was really early days. And we were prototyping.

[00:12:28] But back then, it was, especially for larger organizations, just was much more difficult to move fast enough in the AI space and to do all the work that needs to be done to develop just the one application. But I can tell you that was indeed the goal back then. That's years ago. So I definitely see this being used more and more to tailor and fine-tune a treatment or treatment method for a very specific use case. It's like you see that in cancer treatment, but in imaging, it will be more and more present.

[00:12:56] Just a matter of time and being able to infer what is the right data set that you want to use when you're making a decision. Because there are so many parameters, that's one of the things that we were, that we struggled with. There are like thousands of different parameters that can influence the outcome. How do you choose the right ones to look at? How do you choose, how do you identify the ones that influence the most? So I think there is still work to be done there, but absolutely going in that direction.

[00:13:22] Well, and I think for whatever reason, recently I've been exposed to a lot of aneurysm enlargement data. And I started my, at Boston Scientific, doing AAA stem graphs. So it's an experience I've had. And it's a rule of thumb about when you get to five centimeters, it's the probability of something happening in the next 12 months. And the challenges of the surgery sort of balance each other and it's time to make a decision.

[00:13:46] But now you hear Cleveland Clinic coming back and saying these numbers have some value to it. But a six foot five man compared to a five foot woman, his vessels could be 2.5 to three centimeters normally. So five points. So it's interesting as we're getting this data, we're starting to ask the next level of query, which is are even our markers and decision points we make today, can they be improved upon?

[00:14:15] And so I see that as part of that. What challenges is your organization facing as you move towards this vision? Because you have a lot of moving parts, right? Because it seems like you've made a decision to have a marketplace, an open marketplace to coordinate some of the pieces that you need and have an open system where these pieces. From an AI perspective, yes. From an AI perspective, it's absolutely going to be more and more of an open system.

[00:14:40] But as a solution, it's still a very tightly integrated solution within the hospital environment. Yeah, I think the challenges that we see in the industry translate to what we're trying to solve. Like I said, if you think about the industry and we're not the industry, if you think about the population, right? People get older, you have more of them, and you don't have as many cardiologists and others who actually take care of it.

[00:15:05] You actually get an increase in volume of studies and complexity and a lot more data that gets generated over time. And how do you deal with that? So that's one of the main challenges that we're dealing with. And our solutions are catered towards that space. Like I said, even simple things like making sure that from a workflow perspective, the right study goes to the right person. In many cases, it doesn't happen. And then Dr. X is looking at the study that should have gone to Dr. Y. And then he spends twice more time to get not the best result.

[00:15:33] So even things as simple as that is what we're doing to set integration and other things. But some of the other challenges that you see in the industry is that healthcare is a relatively slow-moving industry. And it's really resistant to change industry. I feel that's like a generational thing. Yeah, we still have a lot of people that have a mindset of how to do things in a certain way.

[00:15:57] And it's really hard to change, A, because healthcare has standards and those standards change slowly for a good reason. But at the same time, it's also being able to adopt through the change and move forward. It's just not an easy population to go about. And from the administration all the way through the care providers, right? They all are in that kind of a cycle. So moving like what we're trying to do now, think about COVID was an interesting trigger, right? At COVID, all of a sudden, people start going to the hospital.

[00:16:28] Get a home. It became a thing. But with that, you also needed to be able to do diagnostic reading from home. How do you do that with an equipment that never was suited to doing that? So working through these things pushed us to develop solutions that today are leveraged. And then today, we're not in that same constraint anymore. But we still have a lot of people that want to work from home, want to read remotely. And you can also distribute the load in a better way.

[00:16:54] So part of our solution goes towards a better streaming solution so that you can actually remote connect from anywhere to our system and do your diagnostic reading. So we are moving with the technology and trying to foresee where the challenges are going to be. But even with all that, take the most recent kind of cloud adoption piece, right? Five, seven years ago, when we spoke to healthcare organizations about cloud, it was, what is that? And really no interest.

[00:17:23] Today, it's the reverse, right? Today, everything that goes around, is this on cloud? Is everything integrated there? So you see the value come together, but it took a long time for that industry to catch up. So that's the other piece that we just need to be cognizant on, that it's not as fast as some other industries. I do think that we see in a hospital setting what I would call the closed-loop systems, meaning you had mentioned that your system is very integrated and tight.

[00:17:48] It's accepting of some of these things because you're knowledgeable of the workflow of your system and the cybersecurity is there, right? Whereas some of the other solutions where I'm tacking on something from an outside system into an existing system, if you're the CIO of a hospital system, every attachment is a cyber risk. So you have this schism that you always have going of interoperability with security. And privacy and so on.

[00:18:15] So I look at companies like yours, having a closed-loop system, are actually capable of advancing faster. And then at the same time, they're offering the app store to allow for creative attachments later at the industry pace that people are willing to accept it. Yeah. But what we do is we maintain, even though we allow those kind of integration, it stays within our ecosystem. So from a security perspective, we do take on ourselves a lot of what the customer would have typically done.

[00:18:45] And because we're able to leverage our cloud partners, we're able to actually increase the level of security that the customer used to have in the past. That's one of the advantages of this transformation, actually. It is a risk and an opportunity at the same time. Now, how do you integrate with electronic health records? That's pretty straightforward. We have standard protocols where these systems connect.

[00:19:08] And we can, so we, the integration is actually all throughout the, from an order generation when a patient comes in all the way through imaging and everything is available, report getting sent out. So we are able to actually show all the relevant information from the EHR in our system so that you don't actually have to go outside of our system.

[00:19:29] That's why I said when a radiologist goes to work on reading a patient, they don't have to go outside our system at all because we bring in all the relevant information from the EHR and then we are able to share that back to the EHR at the end. But all the archiving and everything happens in our system from an imaging and everything else. So it is tightly integrated, but it's nice so that you don't have to continuously look at five different screens and jump in between them. It all comes together in one place for us.

[00:19:59] That's great. So where do you keep current on all the changes that are going on? People like to know where do you go for your information or what websites or what newsletters do you follow? Yeah, I mean, there are a lot of sources online, obviously, but I spend a lot of time talking to customers and talking to our vendors and spending a lot of time of understanding problems and solving them with the partners that we have. I think that's where a lot of that is coming from.

[00:20:27] We do have, for example, in the space of AI, we do have people that are focused in that space and their domain is looking into the technology, into the specific solutions. And my background is in engineering. So I've done a lot of it myself to them. I'm doing a lot less engineering, but that's where I started. So I do understand the technology and where it's coming from pretty well. But yeah, it's really talking a lot about how the things evolve over time by talking to customers.

[00:20:54] There are also, there are a lot of different trade shows where if you really want to look at what's coming next, then RSNAs are on the corner. So we'll be there and look at current and future technology. But we also do a lot of what we call user groups or KO events and so on where we listen and try and get the future trends, not what's going to happen or what the customer needs immediately in six months. But where is it going in five years? Where is the whole need and demand going to be?

[00:21:23] So I spend my time in between these kind of these sources. You're a great straight man because my next question is, what do you see is going to be the next five years? So I see a lot more integration and a lot more automation in how things are done today. So today we have still a few desperate systems in the environment. But I personally expect to see a lot more of that getting integrated into the single platform. So it doesn't have to be the same product, but the integration is going to expand.

[00:21:51] The type of data that we're going to share is going to also expand. So today, for example, digital pathology is really relatively young technology. But having that integrated provides another layer of understanding of what we're looking at. And yeah, I think the future will be a lot more efficient systems producing more accurate data because it's just going to have more sources of data to look at. And they're just going to be smarter in how they interpret that data.

[00:22:20] And again, ideally simplifies the life of the end user because that's our target, making sure that the radiologists, whoever is the clinicians using our systems are able to do that in the best way possible. So that's about efficiency and integration. And I thought I heard I just went check. So are you going into laboratory information management systems and grabbing that imagery too in the future? Today, we don't. Whether or not it will be something that we're going to go to.

[00:22:48] All of the imaging providers will have an integration with genomics data sets and so on because those will become more and more readily available. So we're going to get more layers that help us understand it and from there develop the right treatment, the right solution. So we're not in the necessarily treatment space, but we are going to be able to diagnose the problem in a much more accurate way. What sort of categories do you see emerging in the next five years that don't exist today? Maybe it's 10 years.

[00:23:18] Yeah, I'd be a long one. Categories in what sense? Doing things with this data or adding services because of this system that didn't exist previously. Yeah. So I don't know that there are going to be a lot more sources of data to come through. I'm not sure we're going to invent a new modality that doesn't exist today that's going to produce a new data set.

[00:23:44] But I definitely expect to see more of the data that we have getting integrated. And there is still a lot more data that we're not integrating today. If you think about even your own personal watch probably produces data and information that today we're not really leveraging in understanding your situation. Because when you're going to the doctor, they're not going to ask you, can I connect your watch real quick and download your heart rate from the last 30 days? And we're not doing that, but there is nothing really that stops us from doing that.

[00:24:13] So really looking 10 years down the road, I expect diagnostic phase to be a lot more advanced and honestly, a lot less clinical visits. Because you'll be able to do a lot of connecting the dots in an earlier phase and connecting symptoms in a way that's going to lead you towards a much simpler detection of what you need to do next. I think that's really where we're going to go. So not really in the imaging space, imaging is going to still be their focus on imaging.

[00:24:39] But at large, you'll see more and more integration of the other systems that we have around us that today we're just not integrating. And I think it's going to be interesting too, because I think some of the diagnostic testing as it gets cheaper allows opportunities. Recently, I just did my annual checkup and I said, okay, I'd love to get a heart calcium score. And there was no evidence from the insurance company as to why I should have one. So they did not hit. And so I said, I'm just curious, like how much does one cost?

[00:25:09] It was 200 bucks. So I did it. And so I can imagine that it's going to be really interesting because you're talking about the watch where we're doing our own biometrics on ourselves. It'd be really interesting to watch my calcium score over my lifetime and my blood ox from my watch now and different things like that. Exactly.

[00:25:29] I think that moves the hospital system and health systems in general from acute to having some information to do something with on prevention, which I think is... Exactly right. Yeah. So what else would you like to share with the audience? I think we're in an exciting period, an exciting time in general. I think our technology is there.

[00:25:52] If I think about Agfa, but not just Agfa, our technology is there to be able to offer better solutions. And I think we have the opportunity to keep on doing good for a lot more. Throughout my career, what I cared about throughout the time is what I do matters. What we do matters. When I interact with a physician and they're using my system and they're diagnosing a patient on that, the connection that we have there. So I expect that we would see more and more of that.

[00:26:22] And I don't know, just exciting times in general. I think the technology is booming in the right direction. The adoption is in the right direction. Happy to see the future. And I think that when you think of the long history of Agfa and they started out with, I think, what dies and stains and then moved to photography. There's so many other companies that haven't been fleet of foot. And Philips, I think, is another one that's done a really good job with that.

[00:26:47] Whereas other companies have, I think, of the Polaroid camera being a Boston kid or the Wang word processor. These people that didn't evolve. And next thing, they were a typewriter in a computer world. I think your company's done a very good job with that. So thank you for sharing that insight with us. I appreciate it very much. My pleasure. Thanks for tuning into the Chalk Talk Gym podcast.

[00:27:10] For resources, show notes, and ways to get in touch, visit us at chalktalkgym.com. This podcast is produced by Outcomes Rocket, your healthcare exclusive digital marketing agency.

[00:27:36] Outcomes Rocket exists to help healthcare organizations like yours to maximize their impact and accelerate growth. Visit outcomesrocket.com or text us at 312-224-9945. Thank you.