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

Today on Keynote

(Intro)  where we find ourselves in digital transformation is the pace of innovation to create digital data has outpaced pace of innovation to derive meaning from that digital data.

We are in a situation we have a wealth of data and a poverty of insights.

  My name is Bill Russell. I'm a former CIO for a 16 hospital system and creator of This Week Health, where we are dedicated to transforming healthcare one connection at a time. Our keynote show is designed to share conference level value with you every week. Today's episode is sponsored by Artisite, Dr.

First, Gozo Health, Quantum Health, and Zscaler. Now, let's jump right into the episode.

(Main)  Not being live. I will get us kicked off. All right, it is Keynote and today I'm joined by Shez Partovi, who is the Chief Innovation and Strategy Officer and Chief Business Leader of Enterprise Informatics. Shaz, it's been a while. It's great to see you. It's terrific

to see you, Bill. Thanks for having me on your show here today.

Man, you have you have three titles over there. that's pretty amazing. I we ran into each other couple times when you were at Common Spirit. it Dignity back then or was it Common Spirit?

was Dignity transitioned to Common Spirit. Yeah, it was Chief Digital Officer. I practiced there for a decade, but I was Chief Digital Officer, I think, when you and I were interacting and then became Common Spirit and then I left after that.

Wow,

so you're one of these physicians who has really taken off on the technology side. And so between there and at Philips, you were also at AWS for a while. were you doing at AWS?

I was Digital Officer at Dignity Health and we were really doing digital transformation at Dignity Health and building everything on AWS.

So began to understand the power of cloud in transforming health systems. And then eventually AWS invited and said, look, you're doing this for a health system in the U. S. How'd you like to do this globally? And so I went over there and was worldwide head of healthcare, life sciences, and genomics, and really essentially helping health systems, pharma companies, and genomic companies, and device companies, quite frankly, to really adopt the cloud and transform the way they work globally.

And then, of course, COVID hit in the middle of all that.

Yeah here's what we're going to do today. we're going to talk digital healthcare. We're going to talk gen AI, because you just have to. You can't gloss over it anymore. Intersecting roles of technology and medicine, because as a clinician, you have a great perspective that I want to make sure that we capture for the community.

I want to talk about the quadruple aim a little bit with you. It's been out there for a while. I want to talk progress and how we are making progress. I think the cynical our community is looking at it going, are we making progress? And I think we are making progress in a lot of different areas. We're just not.

Telling those stories, I think, enough although there are some challenges. And I would be remiss if I didn't talk to the chief innovation and strategy officer about futures, like what we can expect in healthcare. But let's start with your current role. Talk to us about what you're doing at Philips the type of conversations you're having and what Philips is hoping to do in partnership with healthcare.

Sure, my

role, I have essentially two roles at Philips. One is really around strategy for the organization. It's a horizontal role. I support, if you will, all the businesses in understanding what is the ideal go to market strategy for the value propositions that we have. And so that If that role involves interacting with my colleagues, whether it's in our imaging, like in our CT, MR, our image guided therapy, our patient monitoring, our personal health, and of course our enterprise informatics.

So there's a horizontal role in shaping the strategy of Philips to deliver value with sustainable impact. And then I have also a business leadership role where I lead the software businesses in Philips globally. So all the software businesses that are In one global team is one that I lead as well. So if you will, both a P& L owner for a business vertically, but also head of strategy innovation horizontally working with the rest of the businesses in Philips.

So I was joking with you earlier that today we find you in Amsterdam. Is that right? Yes. Just outta curiosity, how many frequent flyer miles last year?

Yeah. You don't want , I think I move on like row seven D or something. Yeah, it's it's a flying around a lot. Between, but we have innovation hubs in Shanghai, in Cambridge in the Eindhoven here, and also in Bangalore. And so I fly between those locations quite a bit. And of course we have other manufacturing sites and other things. So that finds me globally moving around quite a bit.

let's start with digital transformation in healthcare.

When we. When we first met, when you were at Dignity, we were talking about digital transformation. The pandemic has accelerated the digital transformation in healthcare. From your perspective, what have been the most significant changes and how are they shaping the future of healthcare delivery?

so the exciting part about the whole digital transformation is, of course, we went a lot from analog to digital.

So that's the meaningful use, the adoption of electronic medical records. And actually, when you look amount of data in healthcare is growing at like 36 percent annually by 2025. And so we have this scenario where There is increasing digital data available, and the problem we have the, so the net effect of that, of course, Bill, is that clinicians and physicians and nurses have to actually, has to now manage that data and to convert that into meaning for better patient care, for the quadrupling, if you will, so improved quality of care, reduced cost of care, and improve the human experience, whether it's the physicians, nurses, or patients.

So this tsunami of data, in some, digital data, in some ways is facing the clinician Overworked, overwhelmed, facing burnout, we have all kinds of pressures and they have now more data to process. So where we find ourselves in digital transformation is the pace of innovation to create digital data has outpaced pace of innovation to derive meaning from that digital data.

We are in a situation we have a wealth of data and a poverty of insights. And so this is where we are today and part of the opportunity, of course, we can talk about AI later. Is that being able to apply new technology. So I think it's exciting to see digital transformation occurring. And it's part of it, at least the United States has been through incentives and in other parts of the world is now beginning to see the value in terms of cost, quality and experience.

But it's only half the story. It's necessary. But not sufficient for, I think, really achieving quadruple AIM outcomes.

The wealth of data, or swimming in data without insights. you think that's part of what has led to physician burnout the United States?

It certainly contributed to it.

The, look, I'll give you my personal real life example. When I was a radiologist, And I would look at a CT of the brain, it was about 24, 25 images, three sheets of film brain windows, bone windows, blood windows, and we'd look at those three sheets. When I was stopping practice, now this is about, years ago, five, six, seven years ago, when I stopped practicing, instead of looking at 25 images for CT of the brain, it was like 2, 500 images.

So you have the same amount of time, And a hundredfold the images, and so you just can't work the way you used to. And of course, that explosion and tsunami of data results in cognitive burden. How, where do you focus? Where do you do? How do you do all that work in the same amount of time? And I'll leave reimbursement and compensation alone for a second, just talk about quality of care.

And time to deliver quality. So yes, I do believe that this tsunami of data has contributed to the mental exhaustion and overburden. There are other factors as well. You have the COVID, you have people retiring, but it is without question one of the vectors. In what has created an environment where physicians, nurses, clinicians are overworked and overwhelmed.

And feel free to talk about Philips and your solutions. I would welcome that. the imaging is one of those areas where AI has been around for a while. So people think like AI just started a year and a half ago with. When we heard about ChatGPT, but the reality is in imaging, we were feeding these images because they're nice from a data standpoint, they're clean data, right?

And so we were training these models and we were seeing significant advancements with AI doing reads and assisting and doing those kinds of things. bit about the progress and where we're at today with regard to AI assisted reads in the imaging space.

Yeah, I think that's exactly your your comment on the fact that it has been around for a while is a great one because we, when you look, for example, Philips, we have 50 or more advanced visualization tools which help the radiologist use AI to actually.

Analyze images, whether it's 3D, whether it's 2D. And so those have been around for a while. We've had a heritage of excellence in, in AI. And in fact, your comment on supporting is a great example is there's a condition called multiple sclerosis. And when I used to read films and MRI films, and MS, you have lesions of the brain, and you have to measure the volume of those lesions.

And so you have to go through every image, And measure X, Y, and Z and add those up to calculate the volume of this lesion, and then it with a prior study to see if the volume lesion is more or less. That's the kind of thing that's perfect for AI, to support the clinician, the radiologist in this case, and we do this for example, at Philips and we've shown that there's a 44 percent increase in accuracy because the machine can do that volumetric calculations with such a repeatable and more accurate way, and also, quite frankly, Radiologists didn't go to medical school and training to measure lesions.

that's not the highest level of certification for the radiologist. So that's a great example where AIA has been around for a while, we've been using it at Philips, and it supports the radiologist in this case to operate at a high level where they don't have to do the menial things that are like measuring lesions that are important to care quality.

So I'm gonna Ask the nerdy question here, which is, as we're implementing this AI, one of the things I think about with imaging is the amount of data and moving that data around. Are we implementing the AI tools where the data resides, or are we actually trying to move this? to the cloud and then back, what's the architecture look like for this and are there challenges in having it scale?

That's the right sort of question. When you look at. Applying artificial intelligence to large image datasets, like imaging, medical imaging is a great example, these are large payloads. And the kind of compute that you need is too expensive to put on premise because you want to use it, it's more cost effective to build a solution in the cloud where you only use the compute power when you need it, and you don't pay for it when you don't need it.

As opposed to buying a piece of hardware, putting it on premise, where it's The cost is there whether you use it or not. And so as you look, as certainly for Philips, when we look at what's most effective for our customers to be able to get the kind of artificial intelligence support they need at the right cost point for them, decreasing the cost of care, We found that it's more effective to solve their problems by moving to the cloud.

So as an example, last November at RSNA, we announced that we moved one of our on-premise solutions, which was our PAC system, which was typically on premise to the cloud. Now it's called HealthSuite imaging. And so now with the images payloads stored in the cloud, we can apply the AI packages, if you will, to help with the workflow using cloud compute power where the cost is only related to the time to use it.

You're exactly right. To really take advantage of all the generative AI as well as sort of neural network technologies, you want to be able to make it cost effective, and for that you want to move to the cloud to really apply compute only when you need it and only pay for when you need Is

there a certain ology, radiology, cardiology, pathology, that is more advanced with regard to its use of AI the others?

When you look at, historically, radiology was the spear's tip. The adoption of digital imaging, like the PACS, was, we started with radiology, was behind it, and then I would say pathology is behind. From a perspective of sheer a number of companies and technology providers that are in the space, radiology is the spear step.

And so I find that and even when you look at internally within Philips, we have the broad, a broad set of AI tools for the radiologist who's trying to do post processing and image analysis and even workflow Flow optimization. So for example, I mentioned HealthSuite imaging. As the images come through, the AI looks at the image and says, oh, there's an anomaly in this and moves it to the top of the queue.

That's a workflow activity. Not even making diagnosis, just putting it face up in front of the radiologist earlier. And so in radiology, I find that there's the leading edge of AI adoption, but cardiology pathology are catching up. Particularly really fast. But I would say that the radiology is leading the charge with cardiac and pathology trying to catch up.

And they will. Speed is

different. we're careful to talk about AI assisted reads and those kind of things. But is there any area where it's like it's so mundane for the technician that they're like, let's just say x ray. Remember the simple x ray? just read them and say, Hey, here's what I see.

What do you see?

Let's just talk about the plain X ray. And there's when I look at artificial intelligence, and even within Philips, there's a dimension where it's around what you're referring to, which is diagnosing things. And those are, those algorithms are FDA over some regulatory, depending on the country, regulatory oversight, you have to meet the requirement and get it registered, and then you can use it for diagnosis.

So let's put that in one bucket, and then there's another bucket of algorithms, which are really around optimization of workflow and process, and both are very important and valuable. I'll give you an example in fact, again, this was recently at RS& A, we have a solution with plain x rays, digital x ray, DXR, digital x ray, and the actual alignment of the x ray to make sure that it's aligned and to, to a point on really mundane activity.

One of the things you want to avoid is having to do retakes, like if the person doing the alignment of the x ray, chest x ray, just is off a bit, and then the film is taken, the x ray beam is exposed to the patient, and then it just turns out that it's not quite centered right, and you have to repeat it.

That's a really great example of AI just saying, yep, nope, this is the better lineup. And so in the actual work of doing the x ray taking, the AI uses the cameras and aligns the x ray beam. in the perfect way to get the perfect shot every time. So that's one aspect and we have lots of examples of those.

And then to your point, we have also have registered algorithms. A good example would be in our ultrasound team. There's a condition called fatty liver. Fatty liver is really hard to diagnose. In ultrasound, you have to be trained. Quite a bit to really get it right, because it could be the machine setting is off, and basically, you just need to know, and it has implications.

It's one of the ones in which it can cause a lot of disease. Two years ago, we registered with the FDA an algorithm that in an ultrasound machine, as you're scanning, it can say, that's fatty liver, and make that diagnosis, and allow the operator to make that call. To your point is, and I wouldn't say that's a menial activity, it's actually a hard activity, and the algorithm supports the user to be able to do it more accurately and consistently.

So Philips certainly does both, and I don't want to underestimate the value of workflow optimization, but also clearly a lot of people talk about diagnosis, but there are both.

So generative AI has taken off in the ethos to the point of mythical proportions to people. And I find myself talking to them about what it can and cannot do and where it fits.

We've talked about a lot about AI and machine learning, computer vision, and those kinds of things. At this point, where does generative AI fit into the work that, Philips is doing and that you're doing at Philips?

We certainly our teams have been really focused on AI for a long time.

And certainly generative AI has been part of the portfolio of activities that we do. We have 500 data scientists and they've been working on AI and generative AI for More than it's been popular, but certainly been there. And a good example of, so we can take a lens of general AI, and maybe I just gave the diagnosis and workflow example.

Now let's look at personas and then map, because the reality is we were looking for impact, not just technology for the sake of technology. So Let's talk about the AI. Let's talk about the administrator in generative AI. Let's talk about the patient in generative AI briefly. And each of these generative AI has a value and Philips is working in each of those cases.

And for example, in the case of, let's start with the administrator, what customers tell us is that I want to understand my entire Workflow in the hospital, and how can you help me create a digital twin that I can understand what's going on in the hospital, what will be my requirement for nursing staff next Friday, what's the time in which my patient will come off the ventilator, what is the, my supply chain requirements for a particular emergency department or the OR.

And these are places where generative AI really help create the kinds of prediction algorithms and prediction Tools that will help an administrator run their organization better. Think of it this way, an administrator will talk to their hospital. digitally. In other words, there'll be enough, there's enough data.

Yeah. That gets back to what you were saying earlier. We have we're swimming in data and that's not only on the clinical side, it's also on the administrative side. And we, to be able to interact naturally in natural language. Would be

fantastic. You're sitting there and says what will my emergency department look like next Friday?

What if I increase my nursing coverage by 15%? if, like this, these scenario discussions will become, you have to think basically like sick pay, right? If fan. You go to sickbay and the, is talking to a, an environment.

And so now let's come to the clinician. And in fact, the sickbay is a great example of what I think generative AI, if you look in the future in terms of being able to talk to the medical record in planning patient care and being able to actually have a conversation. The higher up the patient's chart, so you're, not only is the system so here's a good example.

When you read radiology films, this is a really life example, I used to live this every day. When you go to read an MRI of the brain, for let's say a particular cancer, you want to have all the prior films. You need the prior reports and the prior images. And this body of work for me to pull up the prior studies, pull up the prior reports, and review the patient's chart is important for care, but it's also 10, 15, 20 minutes sometimes.

General AI will pull the relevant priors, the relevant images, the relevant prior radiology reports and conclusions, the relevant chart EMR activities and information, and all that will be generated into a clinical synopsis when the MR of the brain comes up It has, just as you talk to Chad GDB today, it will generate that synopsis and whether it's a textual keyboard based interaction or the radiologist interacting verbally, you will talk to the patient's history as you generate the new interpretation.

That's fascinating. And we were doing that already. We have this example I gave you of pulling up the priors created synopsis for the radiologist is in our labs today, the last frame, I can't ignore the patient. I don't want to ignore the patient. So the same exact conversation we just went through.

Right now, you know the thing that everyone talks about patient portals? So you log on to a patient portal, first of all, you have a report, you're not even sure what it means. You have all this complex medical terms. So health literacy is a huge gap. your record will be an avatar.

Which is age matched to you, that talks to you about your record. You will talk to your record and say yeah, my doctor ordered an MR, why did I get an MR order again? And the avatar will pull from your record in a very personalized way, will educate you. I'm not saying they will make diagnoses, I want to be clear.

I'm talking about the health literacy gap, and you will empower patients because they will understand their own history better. They understand what's going for this entire interaction, just like you and I talked to one of these general AI tools says today. They will be not web browse click.

It will be a conversation with your medical record. These are the vectors of what generative AI can do for us. And we're working on these as well.

  📍   In the ever evolving world of health IT, staying updated isn't just an option. It's essential. Welcome to This Week Health, your daily dose of news, podcasts, and expert commentary.

Designed specifically for healthcare professionals like yourself. Discover the future of health IT news with This Week Health. Our new news aggregation process brings you the most relevant, hand picked stories from the world of health IT. Curated by experts, summarized for clarity, and delivered directly to you.

No more sifting through irrelevant news, just pure, focused content to keep you informed and ahead. Don't be left behind. Start your day with insight at the intersection of technology and healthcare. This Week Health. Where information inspires innovation. 📍 Increase 

 

I

remember when we were and you're old enough to remember this as well, when we were introduced to Google for the first time and we would ask the internet.

Like we would ask it these questions, not with voice, but we would just ask it these questions and it would come back. And I remember being amazed at that point. And generative AI has that same kind of awe inspiring feeling for me as I'm interacting with it. But I remember that Google came out with.

a Google Appliance, and the Google Appliance you used to install in your data center. And then you'd point it at your internal data stores. And it would go and do what it does on the internet, but it would do it with your internal files and that kind of stuff. And then you could ask it questions, and it would return like a Google search.

I don't know if they still have that appliance today, but it was a pretty powerful tool back in the day. Do you think Gen AI is going to be the same kind of thing? Like you plug in the appliance, it goes and It scours your data and then you can start asking questions. Or is this going to be a team of programmers and whatever doing a heavy lift get to this point?

Yeah, so actually I'd love the image you drew because I can maybe give you an example of what we're doing because it really is a terrific analogy that you gave. We talked earlier about the data is becoming more and more liquid, and we hadn't gone to insights yet. For example, Philips has a product called Capsule, and it can liquidate data from more than a thousand different medical devices in a hospital.

Imagine an IV pole has got a readout. Imagine a ventilator has got a readout. Imagine whatever you walk in a hospital, you've got all these Readouts, digital readouts. We can liquidate all that data and put it into a data lake. And then now the opportunity is exactly what you said and we actually do have what we're calling our clinical insights manager.

So you have data liquidation. Which we've done, where the capsule is number one in the U. S. It's everywhere. And then, this data now, is what the customers, I started really saying, the customers are saying, look, I don't want just data, I want insights. And so with Clinical Insights Manager, we're actually helping customers say, what is the meaning behind this data?

And we're actually now bolting on GeneRV AI, so previously, the versions were, it was a, dashboard that shows you trends, this, that, the other, you can create reports. And now in our labs, it's talk to that data exactly as you say. So for Philips, it's a perfect natural story because we are the largest data liquidator through capsule.

And now with the advent of with the, not advent, but with really the acceleration of general AI, the exact conversation you and I had earlier, which is The administrator will talk to the data and the insights will come. The clinician will talk to the data, and the patient insights will come, and in the fullest of time, patients will talk to their data and get education.

So that's exactly what the direction is, and it's sooner than we think. I, clinical Insights manager generate AI tools that are gonna be this year. This is not like we're talking five year because the pace of innovation, is just in general, AI is exploding and accelerating.

Yeah, it's amazing. I downloaded the ChatGPT app about a month ago and then my daughter showed me that there's actually a button on there and you can actually have the kind of conversations you're talking about. You can have with it and it talks back to you. Now, it's basic at this point, and I'm wondering if this kind of tool or if it's going to be A tool that we have to train medically.

If we have to train it on medical terminology, if we have to train it on those kind of things. Or if it's some combination of the two, because Chachi BT already knows how to have a natural language conversation with me but it may not understand the deep medical aspects. Do you think it's gonna be multiple tools working together?

You, so for the foundation models, we'll have to be tuned for healthcare. You are, and we are, for example, doing that as well, but The conversa, the public, the consumer level conversations you're having now, which are GDP are based on the foundation models that are received from open ai.

And when you transpose that to healthcare, I agree that you do need that. And we are doing another as well, which is you need to fine tune it and add additional training so that it's really medical grade. And plus it's not just the understanding the medical terminologies and the medical relations, but also the guardrails.

and this is by the subject is blows wide open as you talk now about regulations and then the sort of the security privacy, but the short version is it isn't the typical consumer grade LLM or foundation model that's really going to be used. Vanilla as is in healthcare.

It will have to be fine tuned, those foundation models for healthcare with medical data. That is a critical step and in addition guardrails, in addition regulatory sort of things, but it won't be just out of the box, a consumer level MLM.

You and I were both in US Healthcare for a long time.

You're still in US healthcare, but you're, you have more of a global view now, and the US healthcare system is a rubric's cube that we've, it's been almost impossible to solve. And then you throw in the complexity of all the different countries, all the different regulations.

We have new AI regulations in the eu. We have burgeoning AI regulations here. I'm sure in Asia there's probably different, and that's just the AI conversation that's going on. You have the FDA in the United States. You have other entities around the world. How does Philips balance that?

How do you look at strategy? Do you end up with regional strategies or do you try to have a global strategy that has regional customizations,

the strategy is. Typically, we would take a position where, we'd take a position where the strategy with business is global, but there are certainly regional tailoring.

And the tailoring is both technology wise, but also geopolitical and regulatory wise. For example, the strategy of saying that what we want to do is help move from data to insights and help reduce burden on physicians and take away friction and give them more time. That's the productivity strategy that's global for us.

On the other hand, that's deployed and exhibits itself very differently in China because of the workflow, because of regulations than in the United States, than in the UK and Europe or in other parts of the world like LATAM. So that's the lens that we take. And of course, anything we deploy in any country meets the regulatory requirements for the country.

So we obviously would never put something in a country that doesn't meet the regulatory requirements. So it is a global strategy. And a global perspective on our purpose and vision, but it gets tailored in every region, both technologically and regulatory wise and workflow wise, because healthcare is always local.

I want to tap into your medical background with, you have a unique background, both medicine and technology how, from a clinician perspective, how is technology being viewed today versus how it was being viewed eight years ago or 10 years ago in the field? Are we making progress with regard to The value that technology is bringing to the at the bedside at the point of care.

I think that

we've passed through some of the hype cycles. Not the generator hype cycle, we're in the middle of that. But when I think of when I used to practice and we talked about cool things, it's the truth. There's a lot more dialogue now on impact. And how does this impact care delivery? How does this impact cost?

How does it impact? And it's the right dialogue, by the way, so I think we've at least, and maybe, one way to look at that is maybe we've become, we as a society, and healthcare in particular has become more cynical. Maybe that's what we're looking at. But I think it's the right thing to say is, it's not just about technology, it's about impacts, which is why As we look at what we build in Philips, it's always focused on what's the impact on care.

You have to do that, and so that is a critical shift I see in most places that I talk to. It's uncommon to talk about, this is really cool, notwithstanding the hype cycle of general AI. But the focus on impact is a really important one that we're seeing more and give you an example.

We have, E patch, which you can monitor monitor patients at home, and monitor post discharge. And the real value of that E patch for monitoring is because it can predict potential atrial fibrillation in the next two weeks. So if you put a patch on a patient, monitors their heart for 24 hours, and then it says in the next, with 80 percent accuracy, 70 percent accuracy, if in the next two weeks they're going to be atrial fibrillation.

This is a significant piece of information and insight to a clinician that's about to discharge a patient at home. that going from the impact of this technology is that I will have a higher accuracy in knowing if the patient's going to have an arrhythmia in the next two weeks is what customers are asking for today, as opposed to, yes, it's cool, we can monitor someone at home.

What's happening is the so what and the now what is the question that's being asked when faced with technology. And that's the right question. Actually, I'm quite thrilled. It's not just what is this and what does it do, but so what, now what? And that trio is more and more being pushed and it's actually an important piece.

certainly for us, that's a key impact is all that we measure on any technology we build.

you're talking to clinicians at this point, clinicians are listening to this show, what's the role those medical professionals play in tech innovation? think 8 to 10 years ago, they felt like we were throwing technology at them, but I think more and more medical professionals are starting to weigh in how do you see them having the most impact on tech innovation in their health systems?

You're

it felt like it was coming over the wire. Maybe at least when I think of when I used to practice, 10 years ago, it's like, what's the new gizmo that comes over the wire and then try to figure out how can I use it? And today, what the role of the clinician is, and I feel this is a really valuable thing, is for really clinical partnerships with tech companies.

This idea, and we do this at Philips all the time, is the idea of Co creating technology with clinical partners is really important because today the pace of technology progress is moving so fast that you need a partner that's focused on tech and then you need a partner that really understands clinical workflow and patient care.

I actually think that, and I'm now going back to my days as Chief Digital Officer I think there are very few healthcare systems that can actually take on developing technology today. The pace of movement is so much, the investment needed, and the talent is so rare that I could probably count on one hand the organization in the U.

S. that I think would be at the scale that they could have their own engineering teams to build their own technology, and that is even, you could probably guess that. But so the role for clinicians is To really partner with tech companies, inside Philips, we know technology, for example, and we have clinical partners.

We recently announced with NYU, for example, it was a press release we did. We're partnering together to optimize nursing workflow and monitoring patients. This is the best of both worlds. Tech company. Tech know how, talent, energy, investments in technology, clinical know how, whether it's nursing, physician, radiologist, come together, build better stuff that has impact.

Besides, for me to prove impact, I need a clinical partner to put it in the environment, to then see the outcomes, to then report on the outcomes, and then claim the impact. That's to me as if I had to say, if I was a clinician today in a health system, I would look to my digital officer or technology colleague, CIO, and say, how do we partner with certain company and to really innovate together?

Because I know there may be still some clinicians that say, hey, CIO. Let's build an engineering team and build it ourselves. It's really hard,

You don't have to convince me that it's really hard. really hard. curious, as you're describing Philips I hear big tech.

I hear medically focused big tech companies, what I hear, and that puts it in the category. of some of the other big techs, Google, AWS, Microsoft, and others. Is that how you view yourself, that you're advancing the industry through really applying technology at that level?

We're certainly a health tech company.

As you pointed out, we if you look at Philips, we have the largest software business in any medtech. company in the world. Our software business is a 1. 5 billion US dollar business globally, and it's the largest one. So and we are getting signals and customers are telling us that they're picking us because we are moving fast, moving agile.

We're actually, we ourselves are partnering with other technology companies as well. And on the software side, It's a very large global software team. It's focused on agile, rapid innovation, partner with customers, partner with clinical partners, test, deploy, repeat, keep going.

So we are really in that team on the software side, doing exactly you called, so the big tech companies would do. Hardware is a bit different. want to acknowledge, we have obviously patient monitoring, our imaging, our IGT, and then we have a personal health division as well, which does consumer directed.

So we do a lot of stuff.

said earlier, we're going to talk about the Quadroop LAME. I think about it, we've talked about a bunch of it. We've talked about the staff experience. and you cursory mentioned automations and even automations of workflow, I think that's going to become more important.

I want to talk about the cost equation. We're entering an election year and in the United States, and we've entered an election year in the United States. that will elevate the conversation, especially around costs personal bankruptcies, all the things that it will elevate.

How is technology being applied to reduce the burden of healthcare on society and the burden of the costs on the patient and the individual? I

think the, we can look at this in a few different ways, and I'll maybe mention two, and then we'll go from there. So first, I want to come back to this story of AI, and there's a saying that I remember, which is, best way to get out of trouble is to stay out of trouble.

That applies to healthcare. It's the best way to reduce, it's if you can avoid an illness or predict and avoid, I talked about that example of atrial fibrillation prediction to avoid a patient being discharged who might have arrhythmia. When you talk about cost, If you can predict, AI offers the nexus of this, where if you can predict the condition and avoid it, it is one of the, of course, that's the better quality of care, no question.

And clearly, it's a better experience for the patient, no question. And the cost equation is in there. So I really do believe that AI has a role to play So that's one frame, and then I'll just mention one other one because I don't want to just make it about that AI tool set, if you will. Access to care is another dimension where cost gets impacted, where if you have people that have no access to care, but through technology you can improve access to care, then again, Early detection, prevention, and reduction of care.

And so that's, of course, for Philips is a huge part of our DNA is sustainability, access to care, and improving the lives of individuals, particularly medically underserved individuals. And so that dimension for technology to improve access to care in underserved communities is another way. And so I do believe that we have to Look at both the dualities, and certainly Philips does, which is, yes, great tech that is in the high acuity hospital, predicts an arrhythmia, stops it, predicts ICU demise, and you can prevent it.

And then people in the sort of communities that with technology can do things that was otherwise not possible, and then reduce costs that way.

closing question is going to be a futures question. It's interesting because the futures question used to be where we were looking out and we were talking about this could happen or this could happen.

It almost feels like the future is happening right now as we're having these conversations. I'm just going to ask you to Put that hat on, looking ahead, what do you see as the next big advancements in healthcare healthcare technology that is going to deliver on the promise of technology in healthcare?

from a philosophical point of view, I think we're at an inflection point with generative AI. And I want to go away from the hype and say something, which is, I think what AI and generative AI will allow is not just to make the hard things easier, but things that we thought was impossible hard.

And healthcare's future is moving things from the impossible category to the hard. That's the leap that's going to happen, not by commoditizing the hard to make it easy. So let me give you a specific example of something that we're working on with the Gates Foundation, which seemingly seemed impossible, but we're at a point where it's now just hard.

And I think that's what's going to change healthcare. For example, when you look at pregnancy, and we have in the United States, for example, even there are like 50 percent of counties have underserved communities for maternity. And there's some in the US of those counties that are underserved, about 30 percent of them are actually considered underserved.

Maturity deserts, because there's just no care, and so the question that with the Gates Foundation together we were asking, which is, how can you have a portable ultrasound so that you can scan mothers in, because the World Health Organization says it should be at least one scan in the first 24 weeks, but ideally more, and that seems impossible in underserved communities, because the sonography, the sonographer training is six months and they're not available in those underserved communities.

So you can't magically just give the ultrasound to someone and say go scan mothers to see if there's a, there's six parameters. How's the baby lying? How big is the head? What's the heartbeat? And you're like, just give it to someone and say, here's a Lumify ultrasound from Philips. Just go scan someone and tell me how they are and if they're high risk pregnancy.

It's an impossible scale problem. We don't have enough sonographers. They can't be everywhere. So it was impossible. It feels impossible. But together with the Gates Foundation, and using Genevieve AI, we built an algorithm where with one Our training, any person does six sweeps of a belly and it gives those six parameters.

Wow. And now you can say, yes, you can go back home. No, you need to drive 300 kilometers or miles. Yeah, I'm nonstop kilometers to the, but the point is that idea of at scale. Now, of course you could have also said, Hey, here's a pilot in some city and here's a pilot, but I'm talking about a scale.

The idea of. Caring for doing the first scan in the 24 weeks, across the United States, across the world, wherever, at scale, was an impossible task. And now it's just hard. And we're doing it. That is how healthcare will change.

that's really exciting. actually, I think that's one of the advantages.

We talked about one of the challenges of having a global company. is essentially that you have to take into account how complex it is. But one of the advantages is, you get to see how healthcare is being delivered around the world. And there are really innovative and creative ways that they're being delivered to remote areas that could be applied to rural healthcare in the United States.

Absolutely. let us not kid ourselves. There are underserved communities in every developed nation in the world. it comes back to the cost, it comes to experience. What's the experience of that mother that can have a scan done in their community, the healthcare extender the community worker which with one hour training has a job and like the cascading effect is the quadrupling.

And that is, was in a previously impossible task at scale, now it's just a hard task at scale. That's how we change it all.

It's exciting. We're going to be talking to our health systems and getting insights. The patient's going to be talking to their medical record. and getting insights. We're going to have physicians who have digital twins and being able to offload some of the tasks, more mundane tasks, and we are solving some of the more complex challenges that are facing humanity.

with regard to maternity and healthy pregnancies and deliveries. Exciting conversation. Shez, we will have to do this more often. I really appreciate your time.

It was great seeing you again and great connecting. Thanks for your time. Thanks for inviting me.  

Thanks for listening to this week's keynote. If you found value, share it with a peer. It's a great chance to discuss and in some cases start a mentoring relationship. One way you can support the show is to subscribe and leave us a rating. it if you could do that. Big thanks to our keynote partners, Artisite, Dr.

First, Gozeo Health, Quantum Health, and Zscaler. You can learn more about them by visiting thisweekhealth. com slash partners. Thanks for listening. That's all for now..