This Week Health

Don't forget to subscribe!

Apple announced Apple Intelligence, their approach to AI. Today we explore what we can learn from their approach in Healthcare.

Transcript

 Today in health, it, I just got done watching the. Apple keynote for this week. It was really interesting, but have some thoughts about it. We'll talk about it a little bit in terms of what they had in it with regard to healthcare, which is almost nothing. And then we'll talk about what I think their introduction of apple intelligence. And their AI framework. Could mean for us, not mean for us, but could provide a roadmap for us. My name is bill Russell.

I'm a former CIO for a 16 hospital system. And creator this week health set of channels and events dedicated to transform healthcare. One connection at a time. We want to thank our show sponsors who are investigating, developing the next generation of health leaders. Notable service now, enterprise health parlance certify health. And Panda health.

Check them out at this week. health.com/today.

Most new stories that we cover on this show, you can find on this, we call it.com/news. Check it out today. Let's see. Oh, one last thing. Share this podcast with a friend or colleague use it as foundation for daily or weekly discussions on the topics that are relevant to you in the industry. Use it as a foundation for mentoring. They can subscribe wherever you listen to podcasts.

All right. Listen to the apple. A worldwide developer conference keynote. Interesting as always for me, because I have so many apple devices in my world from the vision pro to the iPhone, to the Mac book, to the. You name it? We pretty much are an pretty much an apple exclusive family. I remember when I went to it a long time ago. The primary reason I went to it a long time ago is because it was a better experience. And there was a learning curve.

There was a learning curve coming from windows. I felt like I had more control back then of a windows device. Keep in mind, I'm a programmer by trade. I felt like I could get onto, get into and really make the thing, do what I wanted to do on a windows device. But when I made the move, it was really because we're getting to that point where we're doing phones, we're doing. IPads, we're doing Mac books and we're starting to see an ecosystem sort of form. So that's the reason I went there. They they announced all sorts of things around the operating systems today.

A lot of them are let's call them cosmetic. You can you can really customize your phone in a lot of different ways. Again, great consumer kind of experience. It's more fun. It's more personal. Of a device, the more you can personalize it, the more you can move the icons around change colors and do all those kinds of things. They did demonstrate. Just a bunch of capabilities across all the different operating systems. The big announcement, the announcement you're going to be hearing about, and everyone's going to be talking about is apple intelligence. Which everyone thought the foundation was going to be chatty PT, and there's going to be a big announcement of a convergence with chat GBT.

I. I see why people thought that because there was conversations and discussions going on at the end of the day, apple built their own. Intelligence, apple intelligence. They built their own. And they shaped it to what they needed. And then there is a set of APIs that can connect to chat CPT as well as others.

But the one that was announced in the worldwide developer conference keynote. Was chatty P T. So if you're in their intelligence platform and series, still going to be the interface for it. But if you're in that platform and you say, Hey, give me a recipe for whatever. And it determines that information isn't in your personal ecosystem, it's better to go out into the world and look for something. It will say I'm going to ask tattoo PT, that question.

Would you like me to R is it okay if I. Send along the information that you gave me to chats and PT to get an answer. So essentially they're creating a single experience because what they've created in apple intelligence. Is a is a personal intelligence model that looks at your information and stays within the apple ecosystem. They are communicating in every way that they take. Privacy and security seriously, more so than any other company out there.

Quite frankly. But they are really painstakingly through architecture, through marketing, through through their messaging, through their. Tool sets. They are communicating. We value privacy to the point where it's, we're gonna we're going to interact with your photos, but we're never going to upload them.

We're never going to. Apple is never going to take ownership or move that data into their ecosystem. Which is really fascinating to me because I was reading a story as well that I think it was Adobe.

It could be wrong, but I think it was Adobe that was changing. That had changed. There terms and conditions on the software that they're using essentially to say. We're going to use your stuff for training our AI models. They don't come out explicitly and say that, but the wording in there allows them to take all that creative content that's out there and train their AI models. And apple saying no that's yours. We're not going to steal from you in order to train our models. So they've created this personal intelligence model, which will. Act as it, it feels like where they're going is what we expect. We've been talking about the death of the mobile phone and the mobile device.

And it's because we're going to have, we're essentially going to have natural language interaction with devices and be able to ask it to do multiple things. And the application itself is going to fade into the background. This is what we've wanted for years. When technology fades into the background and people interact with it. Just naturally in terms of their movement and the things that they do. That's perfect.

That's where we want to get to and with AI with, especially with generative AI and natural language, understanding. Of things. We're going to ask it to do things. And then it is going to move across a mesh of applications and get us information. For example. I get a text from my, this was their example.

I got a text from my daughter. That says, Hey dad, are we going to do lunch? When you're in Santa Barbara. And I will say. Hey, Siri. Did I get a reservation? Oh, it's going to series actually going to do something right now. Let's shut that off.

This way.

So I'll activate Siri and I'll say, Hey. I'd like to see, have I made an appointment or scheduled anything and it will naturally just go into the calendar and say, nothing's scheduled on that date. Then I'll say Hey Sarah, I'm gonna be in Santa Barbara with my daughter. Is there any chance that there's a restaurant and it will go out and find the restaurant? It'll say, would you like to make a reservation?

I say, yes, make a reservation. It'll put it into my calendar. It'll send a. Or craft a message back to my daughter to say, I've made a reservation at blah, blah, blah, blah, blah. You get the picture. There's this whole ecosystem of applications, but I'm not interacting with those.

I'm not pulling up each one of those applications anymore. I'm interacting with the intelligence platform, the large language model the natural language front end. And it's going across that whole series of applications.

This is what we've been expecting, extracting the applications from the interaction that we have with these devices moving forward. We will have a natural language conversation with a device, with a computing device of some kind, and it will interface with various applications and bring the information back.

Now, again, I want to reiterate what apple has done is created this personal ecosystem that operates on the device. And when it doesn't operate on the device, it's very specific how it interacts with a larger model. In the cloud and brings that information back, only sharing the information that is really relevant or required in order to get that answer. Okay.

So that's their claim. And then they're saying anything outside of that, we'll have a clear deem. Point of demarcation and you will know you are going out to the world. You are sending your information to the world. You're getting information from the world. And so you've left a a safe space and you've gone out into a. Into a public space, if you will, it will be very clear that you've gone into a public computing space. That's the system they're creating as they were talking about this.

I was thinking. Okay, how are we going to do this in healthcare? And as I look at the architecture and healthcare, one of the things that strikes me is. We have a couple of things that are going to prevent us from doing this or doing this anytime soon. One is just the sheer proliferation of applications. When you say we have 900 applications or 600 applications or 1500. Or I think the number I once heard was over 3000 applications. The ability to do this is very. Difficult.

If not impossible. So you have that application portfolio that we need to get under control before we can create anything remotely looking like this. So that's one thing. The second is the data. It's important to be able to utilize the data. And to, To request the data in natural language and to get the data back and natural language.

But in order to do that, the data you're interacting with has to be. Very clear. And so it's a lot easier in some of the applications that apple is dealing with, that it will be in healthcare. Because, when you're asking about a date for a lunch with your daughter, There's a very clear. Data elements associated with them.

When you're asking about a dinner reservation, that's a very clear set of data elements. But the question is how much of healthcare could we turn into really clear set of data elements? And so the data that would enable an environment like this and the applications that are required to create a, an environment like this. Don't exist today. But what it requires moving forward as an architect that can look at that whole thing, not in a app by app basis. But beyond that, remember, this is what we're talking about.

This is what apple introduced here. Is an architecture where natural language into a system, that's able to move across many applications and pull in the information that is relevant and personal to me. If I am a cardiologist and I asked the question of the healthcare ecosystem. I wanted to grab that information and pull it back and present it. To me as a cardiologist.

Hopefully that makes sense. It's a different way of looking at it. So if we're heading in this direction, we have to evaluate our applications differently. But I would say. I. As much as I can see that picture. And I see that picture playing out. And a lot of different areas. And eventually it will play out in healthcare. But with that being said, One of the things we have to do as healthcare leaders. Is, we have to get in front of governance.

We have to learn how to say. No. Or not. Let me rephrase this because that's the problem. We see it as saying no to the organization. Say no, but the reality is we have to educate the healthcare leadership. And then organize the governance. To understand the potential of what is out there. If we will organize it, if we will be disciplined. If we will head towards a vision of the future. But here's what happens if we are undisciplined, we are not going to be able to take advantage of a lot of these advancements in large language models.

We are not going to be able to take advantage. Of some of these dynamics that I'm talking about natural language front end to the entire health system, so that we can schedule appointments via natural language. And we can. We can interact from a clinician perspective from natural language.

We can do notes from a natural language. Yes. We're going to be able to do some of those things because they're going to be application specific, but the broader healthcare ecosystem. When you're talking about 900 applications. That's just going to require it just a different mindset. You can't have 900 applications.

We have to reduce that. That set of applications. We have to start setting a vision for fewer applications as we move forward.

So I think it's interesting to watch this apple keynote. Not from the perspective of what did they roll out for healthcare? Because it didn't really appear like they rolled anything out. For healthcare, they have some personal healthcare personal health kind of things. With the watch and whatnot, but for the most part, there were no. Healthcare specific roll-outs in their keynote today. However, it's interesting to look at how they chose to roll out. Artificial intelligence. As a model.

They decided to create their own ecosystem. Of intelligence so that you could operate within there and know that you were safe and private, that in and of itself is an interesting model for healthcare. Can we use open-source models to create an intelligence model for our health system, for our data, for the things that are within our bubble, and then say, okay, if you're going to go outside of that, boom. Clear delineation clear point of demarcation, letting people know you're going outside of the safe place. And you're requesting information from things that are outside. That's an interesting model in and of itself too. To roll out. Intelligence there's other models for rolling out intelligence, but I think theirs shows a specific focus on privacy. And security. For the for the user, for the for the user of the technology, they are very focused on their privacy. Again, very fascinating. Way that they've approached it.

I think. This will be realized in healthcare. But if I had to guess when this kind of thing will be realized in healthcare, It will be. Past the decade. But we will realize the the the application sets and all the complexity of technology starting to fade away. We will more and more interact with technology through NEC natural language. We the concept of an application might even go away where we ask. A device things and it goes across multiple applications. But we don't really know what it's going across and it's pulling information back to us. Sorta like the internet.

We don't know how many computers it's gone across. We don't know how many routers or ride or hops it's gone across. And it just brings a whole bunch of information back to us. That's how our applications will function in the future. So it might require a different licensing model in a different It definitely will require a different way of thinking about architecture.

That's going to happen in a lot of different areas. That's going to happen in banking. It's going to happen in finance, going to happen in manufacturing. Going to happen in retail. We're going to experience this full on in the next five years across the board. But in healthcare it's at least a decade off just because we have not set the foundation from a governance standpoint, reduction of applications from a data. And data quality standpoint that's required.

So if I were a CIO today, what would I be thinking about? I would be thinking about how to articulate the vision for the future. This was a. A week first attempt from me. I'm going to keep fine tuning this. I'm going to keep socializing it amongst my peers. And getting feedback. From them, I'm going to continue to look at stories and things about how different organizations bring this to bear. Obviously I think. The the AI models around the. EHR offer the most promise in the short term, but in terms of an ecosystem, a natural language front end to an ecosystem of applications that runs a healthcare system. Um, across six to 900 applications.

I think that is a long way off. So anyway, as a CIO, I'm thinking about how to articulate the vision for what this might be, and then what it's going to require to get from here to there. And what it's going to require to get from here to there is fewer applications. Data governance, stronger data governance. It's going to require us to have investments and trials in terms of the large language models and those kinds of things.

So I would. Be looking at it and it also would require us potentially again, potentially. And I believe there's millions of dollars to be saved here. If an organization were to figure out how to move forward with. Developing. A large language model around their own health system. Instead of deciding to pay the $30 per user per month. Kind of thing. So remains to be seen.

What's the right strategy is at this point, but we do know it's fewer applications. We do know that it's a stronger emphasis on data governance. And the, and really recognizing the value of good quality data and better. Data sets across the board. Anyway, that's my rambling for today.

Don't forget. Share this podcast with a friend or colleague. And use that as a foundation for mentoring. Maybe you guys have ideas. I'd love to hear. We want to thank our channel sponsors who are investing in our mission to develop the next generation of health leaders. Notable service now. Enterprise health. Parlance certified health. And 📍 Panda health, check them out at this week.

Health. Dot com slash today. Thanks for listening. That's all for now.

Thank You to Our Show Sponsors

Our Shows

Today In Health IT with Bill Russell

Related Content

1 2 3 268
Transform Healthcare - One Connection at a Time

© Copyright 2024 Health Lyrics All rights reserved