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 the 2 29 podcast.  

It turns out there's a million cardiac catheterizations per year in the United States. Of those only a third of them actually need to have a stent, a procedure. And so think about that. Two thirds could actually be decanted from that pathway and just have the imaging done alone.

My name is Bill Russell. I'm a former health system, CIO, and creator of this Week Health, where our mission is to transform healthcare one connection at a time. Welcome to the 2 29 Podcast where we continue the conversations happening at our events with the leaders who are shaping healthcare.

Let's jump into today's conversation.

 All right. It's the 2 29 podcast and today I am joined by Dr. Lee David Milligan. I don't use your middle name much, but Lee, it's great to have you on the show. 

Bill. Good to see you. What is your middle name?  

It's Joseph William. Joseph Russell. 

Okay. Wj.  

Yeah, because I grew up a good Catholic. I have four names. William Joseph, Thomas Russell. Which, if you think about it, are four first names. I have four first names as a person. So anyway.  Okay. It says Bill to me.  Yeah, that'll work. That'll work. We're gonna do the RSS NA episode, we'll make this an annual event.

You great. Just got back from RSNA I am reading headlines. I did not go up to Chicago. I'm in Florida. Okay. And Chicago is the opposite of where I wanna be right now. It is, uh.

How, how was the weather? Was it pretty cold?  I'm still t falling out my toes as we speak. Yeah, it was cold. And the other challenge that I think a lot of people experienced was actually just getting in the snow came down pretty hard and most flights were either canceled or delayed.

Mine was delayed several times. I was on the tarmac for two hours,  

But still a wildly much larger conference than I think people. Expect. I mean, when the first time I saw the numbers, I was like, wow, there's a lot of people, imaging is a big space. A lot of things going on in this space.

RS NA is pretty much the premier conference for, yeah,  40,

41,000 is what I heard when I was there. Wow. So big number for sure.  

, You're gonna give me your bird's eye view of what was going on. I'm gonna give you the you know, things I'm reading and like, you know, it's outside. Okay. hit the research button and ended up seeing that there's a helium free arms race going around the MRIs.

Nice. Philip ge all addressing the reliance on expensive and scarce liquid helium AI has moved inside the box itself. Generative reports are now, 

sorry. Just before you go on, just for the listeners helium, the reason why helium is needed in imaging is because in the MRI apparatus the magnet gets really hot.

Right, based on the work that it does. And so helium is used as a cooling agent.  

but if you're doing that, you have to install all sorts of, you know, significant hardware around. The device to vent that and all that other stuff associated. Yeah.  

And it fails sometimes. And when it does fail, it's called a quench when it fails and when that happens, it's a very expensive event for the organization.

So the work they're doing around helium free is, I think, really smart work. And I, that's one of the themes. I think it's, yes, we have a lot of, I'll call it pixel related ai. Cool stuff going on. But there's also a lot of really meaningful, important operational stuff happening as well.  

Yeah. So that, I mean that's you know, that's the first trend that I saw.

Second was AI moves inside the box. So. From post-processing software to acquisition hardware. I mean, there, there's just AI is AI is surrounding this space. It's interesting when we go to the 2 29 events now I don't allow the CIOs to talk about ambient. We do talk about ambient, but I, my question is, where are you using AI outside of ambient listening?

And the number one place they go is imaging. it is permeating the imaging space.

 Yeah. And there's a lot to unpack there. In my mind, I kind of break it down into a couple of categories. The one is what I'll call the pixel related ai. So you've got an image and your AI comes in and helps on the diagnostics.

For that, either with the radiologist or pre radiologist. And there's a whole bunch of talk about there that's, that gets most of the press. The second type of AI that's I think really important as a clinician is, I'll call it micro step ai. So the radiologist, when they're doing their workflow, there's a number of kind of key steps along the way.

When we converted from old fashioned, silver based films back in the day to an actual pax, we digitized the images, but we didn't automate the process. The the radiologist had a whole team of people that would help for all those steps. And then somehow, magically when they went to a digital version, those people went away and the poor radiologist was left trying to do their own hanging protocols, trying to decide who to pick next from the, from the, uh, from the queue.

Trying to identify priors and put them up in a way where it's actually meaningful so that you can compare apples to apples. All that stuff kind of went away, and the, I would say the PAX vendors completely dropped the ball on this. This is a great example of where. You know, you built technology you put it in place, but never actually sat with the people to understand what they needed.

It's taken a decade or more for folks to catch up now that those micro steps are being addressed and we can talk more about that. And it's really cool what they're doing. And then the third space I would say that's really important is the operational piece. So just to your point about helium, but more than that, one of the really cool things I think is also within the MRI, the ability to decrease the amount of time in the tube. So previously you go in for a head, MRI, it's a 45 minute adventure, right? It's very claustrophobic, it's loud. They give you these crazy earphones that are like, you know, tubes and you sit there in this kind of awful circumstance.

And so now they are coming up with really cool ways to decrease the amount of time you're in the tube by half. Wow. And the MRI is the most profitable modality that exists within an organization. So it's a really important kind of third leg of the stool.  

It's a profitable modality, but is it a constrained mod modality?

I mean, are we waiting weeks or whatever? So if we're able to reduce the amount of time people are in the tube, are we going to be able to have more throughput in the, on those extremely expensive machines?

 Yes, in a word, yes. For sure you can literally double the amount of patients you can get through in a day if you can decrease that time.

And remember, you've got a number of fixed costs. Your technologist, who's highly paid is fixed costs. Front office staff, electricity for the most part, lights, all that stuff are fixed. And yet you're kind of constrained to your point with this kind of. Really, you know, awkward, long event versus ct.

If you go for a ct, right? It's like a minute and a half and you're out of there. And so getting this one right is gonna be a game changer, I think, for the industry.  

All right, so help me understand the. The workflow, I when you say we, we let go of a whole bunch of people a decade ago and we haven't really replaced those.

, Are you essentially saying that we're starting to replace that with technology? With ai, correct. Correct. We are correct.

 couple of steps just, and there's a whole bunch of steps. I won't go into all of them, but just a couple that come to mind is, you know, you start with the kind of the work.

The work queue, right? The work queue. You might think, well, gosh, you know, you just got a long queue and you grab something from the queue and you go to the next scenario. But there's a lot that goes into that. Right. So if you're a musculoskeletal specialist, an MSK, you probably shouldn't be doing like body, like normal, like abdominal MRIs, right?

You should be focused on knees and elbows and whatnot. Getting that structured correctly, surprisingly, has been a real challenge over the years. So then now have it so that you input your basic demographic information specialty. Things you wanna focus on. And these AI products can actually kind of serve out what is next and what makes the most sense based on load balancing and also your specialty and how much time you have.

That piece is huge. The other one that's huge. Is comparison images. The radiologists have been pulling their hair out over this for over a decade, and so what you really want if you had an MRI in front of you that you're getting ready to read of the head, let's say you'd want the sift and be smart enough to go through the patient's jacket and identify all studies that correspond to the head, not necessarily MRI, right?

It could be a CT scan or what have you, and. Have it be comparative anatomy, not necessarily comparative modality, right? So MRI head, CT head, whatever it might be, and bring it forward in a way where not only is it the right anatomy, but it's also presented at the right level.

So it's the same level, also the same type of slice. Is it a coronal slice? Is it a sagittal slice? Whatever kind of slice it is, and then the right angle. Right, so that you can just look apples to apples and kind of see what you're dealing with. Historically, that has been really difficult to do. Now they're doing it, and not only are they doing that, they're bringing the reports with them.

And these reports, have you ever seen a radiology report? There's probably a thousand words on there. Eight or 10 of those words are probably impactful, right? And so what they're doing now is they're actually bringing up. Not only the report, they're actually hiding the report and they're only showing the relevant information from the report.

And if you click on it, it'll take you to the full report with that piece highlighted. So it's really kind of presenting the information that is actually relevant to the radiologist in the moment they need it. In my mind, I go back to, as you know, I'm an ER doc. I remember back in the day when I had a complicated case, I just walked down the hall and I'd reach out to my buddy Mac and I say, Hey, Mac, I got this really crazy case, and I would be the ai.

I would basically synthesize all the information and I would give him in 30 seconds what he needed to know in order to read it in a way that was helpful for me so that I could actually discharge the patient. If necessary when it was done. And so we've lost a decade, but we're catching up quickly.

 the CIO and me is listening to you talk about these things and I'm like, okay, is this integrated into my PAC system? Is this in, are these integrated into the EHR? Is this a whole bunch of bolt-ons that I'm talking about? I mean, what am I, what are you talking about at this point, am I gonna end up with.

You having to buy, you know, five new systems?

 Great question. So, five years ago I would've answered that, that question very quite differently. It, they were bolt-ons, so they, they recognized there was a problem and then solutions started to come in with these individual solutions. And now, if I would say there's one theme from RSNA this year is that they're taking a enterprise wide.

A holistic approach to the platform. So when you purchase a PACS solution, you're getting your workflow orchestration with it. You're getting your, you're getting your your speech recognition, which corresponds not only to documentation, but they've now flipped the switch and it's truly command and control.

So now you're dictating whatever you're dictating and the report comes up. And instead of having to like identify where in the report you want to put your cursor and then dictate, you just let the report. Fill. And by the way it auto pulls a lot of the information from the actual study itself so that the contrast that's used, the positioning, all that stuff is auto-filled.

And now all you have to do is say what's different than a standard exam. And it automatically identifies where that difference goes and will generate an auto impression. It's incredible.

 Secondary reads and prioritizing reads are two areas that I've heard a lot of discussion around prioritizing reads is interesting to me.

'cause again you talked a little bit about this before and that we're directing the reads to the correct person to do the reads. Mm-hmm. But there's also this whole idea of, we, we don't want. The AI to do the diagnosis. However, if it can do the prelim, let's, instead of secondary reads, let's call it a preliminary read and say this, this read is more important than this.

Read than that, you know, it can prioritize the queue. I mean, that's been something that's really taken hold over the last couple of years. Yes. I mean, I assume we're seeing that become almost standard practice, isn't it?

 think it's expected at this point because, so picture this and I can't, it's not always the same with hospital systems.

But certainly on the, in the telehealth space, they may have a work queues that are 30,000 studies deep. Right. So if in there you have a head CT buried at number 27,500 and there's a bleed in the brain you know, previously it was just understood that they got away, right? It's just one of those, one of those unfortunate, you know, side effects of having high volumes.

Now the AI will actually crawl through. All of the studies identify the high risk, so the pe the brain bleed, et cetera, and bring that to the top for an immediate read that it, it's great for the patient, it's great for the, for the radiologist, it's great for the system. It's just good stuff. 

Where else are we seeing AI start to take hold?

 I would say that the biggest game changer in the AI front to me, in addition to some individual AI products, which we can talk about that I feel strongly about  is in the orchestration of ai.  So. When I was uh, CIO for, for SimonMed, back in the day, we implemented a number of AI products.

And it was hard. It was really hard. It was four and a half months of integration. Once you got it integrated, you had to really tweak it because the sensitivity would be off or specificity is off, and so it would under call or over call, right, your images, and it takes a lot of time. So there's a number of companies now that have created what I'm calling an AI enabling layer.

Which basically means they do the integrations with all of these third party algorithms and companies, and they have a single point of integration into your pacs. Once in your pacs, they can present in, in a way that is not obtrusive, but does identify that there is an AI product at play and. Red, yellow, green.

Is there something wrong with those going on? And they can own the sensitivity specificity, you know, after a while drift, right? They can own that piece of it, but they can integrate it directly into the pacs. So I would say the biggest game changer is that as a former CIOI really hated that part 'cause it took my team away from other projects we were on.

And so now it can kind of free them up. It also allows clinicians to trial things. They can try something for a period of time and if it doesn't fulfill their proforma, they could say, you know what, maybe we don't go in that direction. So I think those enabling layer companies, there's a number of them out there.

There's farham, there's carpal, there's AI doc, there's a whole bunch now, and they're all a little bit different. Some only present their algorithms that they've built. Others like carpal, open it up to anything that's out there that passes their filter. And so, you know, there's different philosophies about which one is best, but either way, it's way better than doing one-offs. 

There's a a resource constraint for people doing reads. Right. So we've known That this is, this has been a challenge that's approaching and I think has pretty much hit us at this point. I assume that was talked about pretty significantly or addressed with a lot of different approaches.

 Yes. and I would say that there's a couple different takes on this. Yes, it is true that radiologists are really taxed. There's 45,000 radiologists in the United States today, 500 to 600 million studies per year in the United States and, you know, and growing. And, but if the structure of the workflow is not set up great.

there's a chance that radiologist hates ai. because it just adds work and it makes it more complicated when the AI says, Hey, there's an abnormality, and the radiologist says, no there isn't. Right?

Right.

So it can get complicated pretty quick for those folks. And so, yes, that is a real issue. The other issue that is actually getting more press than that is the lack of technologists.

 Right.

Yeah. So the MRI technologist  who's acquiring the images, and there's a number of companies that have created tele-technology scenarios. So you can have a single technologist who can actually obtain images from three separate locations. Right. And they have one person like position the patient, they can talk to the patient remotely and they can work all the buttons for the MRI machine remotely.

And they actually have products that do that. I know. Deep Health has tech Live and there's a few other ones out there as well.  

That's interesting. I would've never, I would've never imagined that, winners from the show. I mean, you wanna talk about some products and we can do that 'cause we don't have a lot of uh, it wouldn't matter.

But uh, we don't have a lot of, of imaging sponsors. So, uh, love to hear love, love to hear who the who, the winners or the people that got you to think think a little differently about something.  

Now we're diving specific into specific areas that I think are the most impactful.

So, one that really pops for me is a company called HeartFlow. So heart flow. What they do is they basically do imaging of the heart and the coronary arteries that feed the muscle of the heart. And so if somebody has, there's a chance they may have, you know, a blockage of some sort.

Historically, that person went through a number of tests to figure out what's going. I start with an EKG. You might do like a treadmill test. After that it might be something like a nuclear imaging study like a csta mibi. And eventually you may go to cardiac catheterization. Cardiac catheterization, it's a big deal.

I mean, they're putting a big needle in your groin or your wrist. They're snaking a guide wire into your heart. They're releasing dye and they're taking pictures of it and. It has a risk with it. One in a thousand patients will actually experience a stroke from the actual procedure, and it's expensive.

It's in the hospital. You've got these highly paid people, a lot of equipment and whatnot, 20, $25,000 a pop to do that. What HeartFlow has done, and also another company called Clearly is doing as well. I know heartful a little bit better, but they're both doing similar approaches, which is they are taking images of the heart of the vessels, and they're doing two things that are pretty unique.

One is if there's a plaque, there's a blockage. They can actually measure the pressure. Inside the vessel before and after the blockage. That's called the fractional flow reserve or FFR. That number's important because it helps determine whether you need to actually do something about that plaque.

The second thing they can do, and by the way, that's FDA approved and it's CMS reimbursed. The second thing they can do, which is really fascinating, is they can look at the plaque and they can characterize the nature of the plaque, so they can say, is that a stable plaque or an unstable plaque? Back in the day, we used to think that when people had heart disease, they would have a plaque and it would just slowly build up until it got to the point where it just blocked everything off and then that was a heart attack.

Well, now we know that isn't how it works. Some plaques are actually quite stable. They can be big but stable. Other ones are unstable, and when they're unstable, they break off, and when they break off, it leaves a raw surface and then plate lifted here, and then you have a clot and then everything goes haywire.

And so they now can look at that plaque and determine whether it's stable or unstable. That is also FDA approved and CMS reimbursed in my opinion. In the next five years, we're gonna go to a scenario where instead of going to cardiac catheterization pretty early upstream in the workup, they're gonna go right to heart flow.

Just get an answer, get your pressures look for plaques, and have an idea whether you need to do anything or not. And it's gonna save money, a ton of money, and it's gonna save people from having strokes.

 that's pretty amazing. Good stuff. So is that commonly available at this point?

I mean, yeah. So, okay.

That's,  yeah. The challenge is that, you know, money has flown historically to the cardiac catheterization lab. And so, you know, it's one thing to develop the technology and have it in place. It's another thing to get institutions thinking differently about how to do this. I will say that because the evidence is so strong that there's good studies now that look at the accuracy, literally comparing the pressures from a cardiac catheterization and the imaging.

I think the pressure is on right now to actually move it into formal care process pathways. And as that happens, it will certainly change.

 I mean, you talked about the risk and you talked about the cost. Is the cost dramatically less?  

Oh, yeah. It's in the order of, it's a magnitude of 10. 

Wow.

Yeah. So you could just put it this into a, you know, I'm. 55 years old. We're just gonna go ahead and do this now. 'cause I'm 55 years old. Yeah. Fam, family history of heart. Let's go ahead and take a look at this.  

Yeah, absolutely. And remember with no risk of stroke and it got, you get both the pressure information and plaque information.

It turns out there's a million cardiac catheterizations per year in the United States. Of those only a third of them actually need to have a stent, a procedure. And so think about that. Two thirds could actually be decanted from that, from that pathway and just have the imaging done alone.  

But it does cannibalize some revenue.

And that's why this will be to, that's  right. That's right. Yeah. No I, yeah, I can see that.  Just the reality of the, of the uh, mechanism that is healthcare. It's like even though we see how clearly this is really gonna be good and the studies, it just takes some time to, to work through the system.

 The other one that comes to mind, and there's a whole bunch, so I'm just cherry picking here a little bit. So, excuse my cherry picking here, but there's a company called AZ Med. And I know this one pretty well 'cause I implemented their software when I was at Simon Med, but they just do x-rays, but they do x-rays really well.

And they started with just long bone x-rays and they added axial skeleton, so cervical spine, thoracic spine and lumbar spine. And then they went to chest x-ray and chest x-ray. It sounds simple, but because it's a two-dimensional picture and it's a three-dimensional object, it can be very tricky.

So they've gotten really good at chest x-ray and long bones and axial skeleton that can take a whole bunch of work off the plate of radiologists if that is in place for us. When we implemented that, we decreased our turnaround times for x-rays by 36 hours, like overnight.

 now again, I wanna simplify this, and I'm not a doctor, so, you know, give me a little grace here, but Sure.

X-rays are the simplest of all images for AI to read? Aren't, are they not?  

No, they're not.  

Why can that, how can that not be Because

 they can't see around corners? If you have a CT scan or an MRI, it, it really is a a three-dimensional, right? So you can see image.  I got

it.  

The x-rays a two dimensional image, so you're constantly chasing shadows.

And so it can be it can be very difficult. So AI struggles with x-rays. Well, AI struggles with everything in the beginning, but it has to learn patterns. So, like anything else, it learn, it learns patterns and pattern recognition. What I will tell you is the FDA doesn't approve AI for a specific anatomic part or modality.

It approves it by diagnosis. So, for example, if a chest x-ray is. You know, integrated into the workflow, into your pacs, and it says negative. The question you have to ask is negative for what? Because right now it's only approved for like, you know, 10 or 12 different things. So you have to recognize that if it says negative, that's fine.

But what did it not identify? So there's no collapsed lung, there's no pneumonia, there's no this, there's no that. There's fine, but just remember it's negative for the things it's approved to to identify.  

I'm gonna give you a lot of latitude here 'cause you walked the floor and you talked to a lot of different people.

I mean, what's I got my

steps in?

Yeah. What's what else should we talk about that I'm not seeing in the press releases or,

 I'll give you one that comes to mind. It's a little bit off the beaten path, but kind of Cool. So, let me first start with the problem that they're solving.

So when patients have been diagnosed with something that requires a surgical intervention sometimes they go doctor shopping and, you know, we don't have a great transparent financials in healthcare. So they're not looking at price. They're saying to themselves, somebody's gonna be inside my body doing work.

Like, who can I trust? Right? And it turns out that some doctors are better at you know, interacting with patients in a way they're comfortable. And so if you look at the total number of patients that they ultimately spend time with, there's a fraction of them that ultimately will go someplace else, right?

And so if you look at doctor a, it might be maybe 80% capture rate. Dr. B might be 60% capture rate. Dr. C might be the best on the planet, but he has the bedside manner of a turtle and his capture rate is 20%. You're like, what is going on with this guy? Well, it turns out that one of the reasons that patients cite that they don't want to go to the doctor is they couldn't understand what the problem was, and that person couldn't explain it well.

And part of the reason is because we're frequently showing patients a CAT scan or an MRI which they haven't been trained to read, and we're pointing to something as I see that thing there, that's what I'm gonna go after. Right. Whether it's a neurosurgeon or an orthopedic surgeon, whatever it is. So there's a company called Avatar Medical that came out with a software program that they've been working on for years that takes the original image.

and converts it into a simplified image. And they've re, they've recently partnered with Barco that you may be aware of. They do the screens right for, primarily for mammography. They started out, now they do everything and they, now they've created a three dimensional image and you don't even need to wear.

Head gloss. So you actually just walk up it, there's a bar on top. It sees you, it locks in on you, and it creates a three dimensional image that's simplified compared to the CAT scan or the MRI that's originally placed there. It is a game changer in my opinion because when I walked up there and looked at it, I thought to myself, you could be anybody and you would understand what the issue is.

So that was one little bit off the beaten path I thought you might find interesting.  

Man, that's that's pretty wild. I, so you, I mean, you've been to hims, you've been to Vibe, you've been to these other conferences. I mean, what's the distinction of this one over, over those, for those people who haven't gone to it?

 They drink more here.  It's the middle winter in Chicago. Of course they do. Yes, exactly.   But it's the same feel. Right? A big floor. I mean, you spoke at in one of the booths, so there's speaking in the booth, there's probably some training that's going on as well. 

Yes. And I will say that what one of the things that's a little bit different than other conferences I've gone to in the past, there's a lot of people who go, because they're trying to learn about imaging. In a way that informs them for a future decision they're going to make. So CIOs will show up with their teams, literally with the PACS admin, chief Operating Officer, with the VP of applications walking the floor, looking to learn more about this because they recognize that, you know, and I can say this as a former CIO.

I dabbled in imaging, right? I knew enough to be able to help support the storage, et cetera, et cetera. But I didn't know a ton about it. Kinda like I do now. And so I feel like it, this is an opportunity for them to really have almost like a bootcamp on, on what's available. The challenge, of course, is there's a ton of vaporware and just like any field and everybody's telling you something different.

So it's hard to get clarity, I think, for folks because there's a lot of noise. It feels very chaotic when you're there. But they are there, walking the floor trying to learn. And that to me was a little bit different than other places I've gone.  

For those people who have seen you on the show before, give us an update of what you're doing today and why we're talking to you about imaging. 

Yeah, happy to. So I founded a company a few years ago called Asbury Health Tech Partners, which focuses exclusively on medical imaging. I did this right after I finished as CIO for a very large ambulatory imaging company across the United States. And we really help hospital systems and ambulatory and telehealth companies pick software in this space.

And we, you know, we started with pacs and Riz and some of the ancillaries, et cetera. But since that time, in addition to implementation, we've been focused on really kind of helping companies develop some of their software, kind of recognizing where some of the gaps are and helping inform them on where they could steer the ship, both on the back end in terms of architecture and the front end in terms of ux.

So it's been a ton of fun. The team has grown. Our kind of, our footprint has grown in this space as well. I kind of think myself as I kind of put myself in the shoes of myself a couple years ago as CIO for a hospital health system, and I try to kind of convey some of the things I wish I knew back then as I'm navigating this.

And I also kind of think of myself as you ever seen that movie at Princess Bride? . 

Yes.

 So you remember when Wesley and Buttercup. Came up against the fire swamp, right? Yes. I, in my, on my best days, I think of myself as Wesley trying to get through the fire swamp because there's so much noise out there, and I, more than anything, I try to clarify reality and really kind of tee up the true value proposition so people can make good informed decisions about what's best for their company. 

What would Dr. Milligan today be saying to CIO, Dr. Milligan? Back at Asante.  

Yeah. I would've said spend a morning with your radiologist, buy him or her a cup of coffee and just watch what they go through. Take and take some notes.  

I, I talked to you after one of those experiences.

Is it so somebody brought you in to look at their environment? Yep. I talked to you after one of those environments. They're like. Oh my gosh. Like if they just spent five minutes with the radiologist, they have so many challenges and stuff that they wanted to communicate, but they're, you know, they were broadcasting but no one was listening. 

That's right. And what's interesting about it, we frequently, you know, complain about issues and problems we have, but in my mind it, it's infrequent when you see a lot of complaints where the solution is so obvious.  

Right. And a bunch of, it wasn't like radiology kind of stuff, it was like basic networking stuff.

That's right. Yeah. And so I think that's one of the things that's gonna differentiate you. First of all, physicians. Second of all former CIO so you understand the technology and then. Third having done the ambulatory imaging, really understanding the imaging space, I think gives you a great perspective.

Not that this is an ad for you, but since you were kind enough to give me the oh, of course the rundown of what uh, what happened at RS NA and I would uh, one of these days I'm gonna join you on that floor. And 'cause it is one of the cooler spaces from a technology. Perspective, I believe  

I'm gonna lobby to have you join me.

We'll partner in it next year. Alright. And we'll divide and conquer 'cause there's so much to do.  Yeah, I can imagine.

Although I'm gonna have to buy a new wardrobe. I have a feeling. So we'll see. See what happens. Lee, thanks for coming on the show, really appreciate it.

All right, thanks Bill. Take care.

 Thanks for listening to the 2 29 podcast. The best conversations don't end when the event does. They continue here with our community of healthcare leaders. Join us by subscribing at this week health.com/subscribe.

If you have a conversation, that's too good not to share. Reach out. Also, check out our events on the 2 29 project.com website. Share this episode with a peer. It's how we grow our network, increase our collective knowledge and transform healthcare together. Thanks for listening. That's all for now.