Speaker 1: First up, we'll hear from Christian Hardal, EMEA Healthcare Leader at SAS. His focus is on health innovation and the increased convergence between healthcare and other industries. As an expert advisor in health data and AI, he works to improve performance, health outcomes, and patient safety for hospitals and health institutions in the EMEA AP region. And today he'll show us how to simplify health data management through SAS Health. Christian, thanks for being here. Tell us just a little bit about SAS and healthcare.

Speaker 2: Well, we almost have a 50-year history in data and analytics, so where to start, right? SAS offers healthcare customers an end-to-end platform that operates on data and analytics, and it operates on-prem or in the cloud like Azure. Our customers use SAS to predict future health outcomes, investigate health fraud, optimize across the capacity management and workforce, and most importantly, we try to improve the health of individuals and populations.

Speaker 1: Fantastic. Thank you. Can you maybe double-click on SAS in healthcare a little bit for us?

Speaker 2: Sure. Last year, in September, we released the first version of SAS Health Solution, which is a product embedded into the Viya platform, so you have all your health IP inside of the analytics platform. What it does is that it improves the health data management across the entire value chain in healthcare, and it supports the entire analytical lifecycle, so data management, analytics, and the deployment of your analytics. What we are trying to do is to automate the ingestion process of your health data management and the industry standard data like FHIR. So basically, taking that data, ingesting it into the engine, automap it into a common data model, and that data model is optimized for healthcare dashboarding and the development of future AI algorithms.

Speaker 1: Great. Maybe you could talk a little bit about no-code, low-code environments and some of the benefits that you're seeing pull through.

Speaker 2: Yeah. So, the great thing about low-code, no-code is that it targets a new audience. So, of course, we provide a code editor for hardcore data scientists where they can do their open source and their SAS integrations, but we also offer this low-code, no-code environment for business users and clinicians, so it makes AI available to a larger audience. So, that's the idea behind that.

Speaker 1: Incredibly interesting. Thank you. What happens after the data ingestion with SAS Health and the linkage into Azure FHIR APIs?

Speaker 2: So, basically, it's accelerating health innovation. That's the goal of it, right? So, we have optimized the process and accelerated it from data management to the analytics to the health outcomes. That's the basic idea. We are also working on GNI super cohorts and out-of-the-box insight dashboards. So, basically, accelerating, again, the entire process for clinicians and healthcare staff to get value out of the data that they're working on each day. Our ambition is to improve patient care and patient lives with the use of data and analytics. So, basically, we know that that means that we need to work closely with our customers and partners like Patricia to get this going because it's not an easy task. And if you get to that space where you're collaborating with your customers and partners, then you have a trusted environment and then you can really start innovating and create these progressive solutions that will actually change how we drive healthcare today.

Speaker 1: It's super interesting hearing where you're investing and the direction of travel. Can you talk a little bit about the role of AR you see in medical decision-making going forward?

Speaker 2: Yeah, it's really a question of speed, accessibility, and then I think most importantly, it's based on real-world patients. So, many, many patients, and it's not longer up to the individual doctor and his gut feeling or her gut feeling on how this patient should be treated. So, what we give is actually a second opinion so that the doctor will be able to rely on all of that information in just seconds. And the computer is super fast at providing information and looking at patterns. So, that's really the strength in this, from my opinion.

Speaker 1: Excellent. Why don't we hear from you, Patricia, to bring it to life with a bit of a real-world example for us?

Speaker 3: Great, thank you so much. So, every 60 seconds, the life of an expectant mom or her baby is lost due to a pregnancy complication known as preeclampsia. So, the AI Primi team is on a mission to change this by combining patented biomarkers with AI to transform the diagnosis and care of preeclampsia. So, AI Primi is a groundbreaking new test that uses advanced AI algorithms to analyse both patient clinical and blood biomarker data. And it will really help clinicians reduce the competing risks for both moms' and babies' lives. And the whole idea is that it will save lives. Really, what it is, is to make decisions in real time from whether to send the mom home or whether to deliver that baby. So, really important clinical decisions in real time that will absolutely save lives.

Speaker 1: Excellent. Can you tell us a little bit about why you chose to go on this journey with SAS and Azure?

Speaker 3: Absolutely. So, we are a team of biomedical scientists and expert clinicians, and we are not data scientists at all. When we first discovered our blood-based biomarkers a few years ago, we worked with expert statisticians to understand the value of our biomarkers but also to integrate them with clinical data. And the results looked really promising. So, bringing this then onto an analytics platform like SAS Viya that we're using in the Microsoft Azure Cloud, we then were able to move much, much faster and we could then make data-driven decisions on our algorithm. So, we were able to explore and understand our data and also understand it using the open-source models that we were used to. I think the most important thing for us was that we could concentrate on gaining high-quality curated patient data and we didn't have to worry about looking at the coding or how we were going to run our algorithms. That was the most important thing because really what we wanted to do was create really rich data sets that we could define the best algorithm possible that we could actually save lives. Really great.

Speaker 1: Thank you so much for the overview. Can you talk a little bit about the commercialization plans for AI Primi?

Speaker 3: So, the collective mission of the AI Primi team is to get our test to everyone who needs it across the world because we really do believe we can save lives. So, to facilitate this mission, we really needed to think about being enterprise-ready and I suppose being ready to scale almost from day one. So, if we were going to get this into widespread clinical practice, we really felt that we needed the best in class analytics combined with the best cloud providers that we could make the path from our lab into the hospital environment absolutely seamless. So, once this research phase that we're in now, once that research phase is complete, we do believe that the combination that we're using of SAS Viya with Microsoft Azure, we really do believe that that will be able to help us get our model into production immediately and that we're able to do this by accessing, receiving, and ingesting data in, again, highly secure, in a really defined and a really regulated manner because that's the most important thing we do believe for healthcare. And if we can do this in the future, then we'll be able to scale really quickly and we'll be able to get this into every hospital across the world because that's what we want to do. We want to get this test to everybody who needs it because at the end of the day, we, as a team, really believe that we will save lives.

Speaker 1: Well, thank you so much for that incredible walkthrough and showing the impact that SAS Viya is having from a business perspective and a patient perspective. I just have a few follow-up questions, if that's okay. Christian, maybe you could give us an example of two tasks you think that could be automated best with artificial intelligence and maybe one that you think humans could do best.

Speaker 2: Sure. I will start with humans because humans, and especially humans in healthcare, are like really caring people. And we still need a human in the loop when we're talking AI. So basically, we need those clinicians to look at the patient and considering whatever prediction or outcome they are being delivered by the AI, and they need to take a real look at the patient and say, okay, is this the right decision or not for this patient in his or her current situation? So that's the first thing that's still very, very important. But when it comes to it, machines are really, really quick at looking up data, looking at the history of a patient and combining it with new acute information from the patient. So we can really rely now on a machine gathering all of this information to you as a clinician, finding patterns, coming with predictions, and then you have like your new best friend in that because he will deliver or the machine will deliver some answers for you as a clinician where you can actually do some fact-based decisions upon. But still, we need that human in the loop to say, is it right or is it wrong? And is this the right direction for the patient?

Speaker 1: Fantastic. Thanks so much. With new technology comes new challenges. Can you maybe share a little bit about what you're seeing from a SaaS perspective and the challenges ahead? Yeah, sure.

Speaker 2: So the two major challenges is first with the AI and then being compliant with regulations like AI Act. So basically, SaaS Viya is giving you both. So basically, we can and we are doing everything we can to make AI trustworthy across the entire analytical lifecycle. So that comes from bias detection. It comes from how you evaluate and validate the models. It comes on how to you actually represent the results to clinician because if it's a black box model, they probably won't trust that if it fails once or twice, then you simply lose your commitment to those models. So basically, it's about bringing transparency into that and that's what we do. When it comes to the regulations, we are having a platform that is giving you all the tools to be compliant with the regulators. So you can document all your results, your data, your lineage, etc. I love that.

Speaker 1: Any customer or real-world stories that you may be able to use to bring it to life for those using Viya? Yeah, sure. I would say in the last couple of years,

Speaker 2: the majority of success has been in the research space. So we have seen lots of projects driving innovation on SAS Viya. We have seen lots of projects actually go into scientific papers and being published in papers. And we have seen nice work like Patricia is doing. And I think we're seeing now the next wave is coming. So basically, what we are in front of now is for all of these data, in front of now, is for all of these nice research projects to go into deployment. So how do they take the work that Patricia has been doing? How do you take that and scale it up to an enterprise level? That's really where we are at right now. So that's what we're trying to facilitate and that's what we're working on in a partnership together with Patricia and together with you guys.

Speaker 1: It's been a super dialogue. Is there anything else maybe before we close out that you'd like people to know about SAS?

Speaker 2: I would just like people to follow the QR code on the screen to go and see what SAS Health is, what the Viya platform enables our customers to do. Also in other industries because we're not only serving healthcare. So that's really a go-to action for the audience. I would also say that there's also a place where you can find an ebook we co-created last year on health analytics in the cloud. So there's a lot of good content if you follow that QR code. Great.

Speaker 1: Listen, thank you both so much for sharing and for being here with us. Appreciate your time. You're welcome.