Today on This Week Health.

Talk to me a little bit about your dataset. Tell me about the features that were selected. Tell me about the model build. What type of model is it? What the heck is it? And then what are the outcomes? So clinically, I want to know what the data looked like, what features were selected. And I want to know everything from start to finish.

It's Newsday. My name is Bill Russell. I'm a former CIO for a 16 hospital system and creator of This Week Health, a channel dedicated to keeping health IT staff current and engaged. Special thanks to CrowdStrike, Proofpoint, Clearsense, MEDITECH, Cedars-Sinai Accelerator, Talkdesk and DrFirst who are our Newsday show sponsors for investing in our mission to develop the next generation of health 📍 leaders.

I came from a really good session. The NIST framework for health care.

NIST framework for security.

Yep, absolutely. So the, kind of the combination of the cyber side, as well as the privacy side and what this has been doing for the last couple of years and how it applies to healthcare. It's a real nice framework that whether you're a large organization or a smaller organization, you can follow it.

Yeah. We, I mean, we adopted the NIST framework back in oh my gosh. It's been around for a while. This isn't a new thing it's been around for a while.

No. They got a huge team on this of four people.

Fantastic. All right. So it's interesting. You talk about security. I find we talk an awful lot about security, but not that much about privacy.

Privacy is extremely important. Security, privacy is trust. Right.

Right. Yep. Charles, we got a lot to talk about. Absolutely. I got my computer in front of me because I wanted to see what people are actually writing about because we got so many press releases here and I'm not sure. I mean, literally about almost 200 emails a day with press releases. Things that were going on, people coordinate around this. And I'm not sure which ones were worth anything but I figured Scott Becker would figure it out for me and others would as well. But before we get there.

Let them do all the heavy lifting.

Do some of the heavy lifting. What are your thoughts? I was talking to DREX last week and I predicted somewhere around 10,000 people. And this is probably in the 20.

Mid twenties. Yeah. And about what a thousand vendors, something like that.

Yeah. The hall is a pretty extensive.

You know what's cool about this year versus others? With only a thousand vendors, they're well spaced. It's clean. You're not bumping into folks and you got time to look and all the important ones there.

So I stopped, I walked through there tonight around some of the booths. Some of the insights were more activity than we. A lot of users are here. So people were saying, Hey, people were stopping into to actually look at the demos field of demos and whatnot.

Probably the most insightful thing I heard from somebody was this is pretty similar to previous years. I said, well are you seeing the CIO's here? I said, look, let's, let's be honest, CHIME happened. Then we would see the CIO's for like the first half of the first day. And then they'd be gone anyway. Or they'd be dragged into meetings.

So we wouldn't see them. And most of the time this person said, most of the time we were spending our time with influencers, not necessarily buyers. And that's fine. That's good for them. I mean, they're working with people who are going to be a part of the decision.

Yeah. The folks at the keyboard. Yeah. That's what you have here. The people that make it all happen.

Did you get to any of the sessions? I did not. I've done 16 interviews. I've done.

Only the ones that I moderated.

Oh, so you moderated. What else did you moderate?

So I, I moderated another one on kind of the conversational agent space in healthcare. So we're now using conversational agent for appointment setting, for general information and so forth.

Yeah. I mean, that's, that's an exciting space. You and I have talked about that before. I think there's all we can wait for healthcare, we can pick up tone, we can tie it to automation. We can, there's an awful lot of things we can do.

What I really like about it is we've gotten to the point where patients don't realize that they're talking to a machine, so basically we're passing the turning test.

Yeah if people want to have a conversation Charles and I have this conversation all the time. So if you want to join in on this conversation just shoot me a note. We'll probably just do like a public zoom and, and talk to people about it. Cause I think it's, it's a fun space and every time I talked to you, it seems like we're advancing in it.

We are. We've got to the point where we're using graph technology. So basically the technology can scrape organizations websites. All their data assets instead of canned responses, their actual responses based on knowledge. So it's pretty cool where we're going with that.

All right, let's hit some news stories. The new regulation would send hospital pandemic data back to the CDC, despite its difficulties collecting hospital data throughout the pandemic. The CDC planes to add data collection back into its repertoire, according to our proposed regular.

Reported by Bloomberg on March 14th, the regulation which Bloomberg obtained from CMS draft would require hospitals participating in Medicare and Medicaid, all hospitals pretty much to report data to the CDC is infection tracker, the national healthcare safety network. Although some supporters of the proposed changes say it will help the CDC increase the ease with which it analyzes and receives data.

So let's it straight before we travel down that down this road. I think this is, I think it is gonna be great for retrospective, but I think it's really setting us up for, for the next one, if you will. I think that's probably the majority of the purpose behind it.

So you think some of this is still aimed at getting a picture for what happened over the last few years

Absolutely. And then setting us up for next time.

Because if we can do it in a retrospective we'll have to create those connections. Create the data data mapping and other things. Isn't the data mapping pretty well set at this point, I would, I would think it's pretty clear. We want these elements.

Sure. So let's, let's say we've got the framework in place, but let's get what we need to in place so that it can be done in real time. Or if we'd done it for 12 hours, 24 hours, whatever that needs to look like. So people can actually react on data now. Not data happened 12 hours, 24 hours, a week, three months, whatever it is. Yeah.

Yeah, so that that's all going to be about implementation. It's going to be about how they spell that out and the clarity they have within the within the law itself. I think we'll, we'll set the stage for.

Let's see if they can open it up for academic researchers as well. Well Epic's got a big one, right?

Yeah. So Epic launch is SAS aimed at independent medical group. So I assume so here's what they're saying essentially. If, if somebody can't find a community connect partner, Epic will essentially have a platform that is community connect for everybody. I mean, I don't think this is going to be a platform that you can highly customize and do those kinds of things to.

No but If I'm a onesy, twosy or three or four know practictioner you know absolutely. I can now especially if in my in my geo area there's Epic hospitals all around me. I can now participate, right?

Right. So a shared environment called garden plot is intended to streamline access to Epics tools for providers with minimal overhead. Right? So you, you roll this out. It's probably on, on a foundation build and it's probably a pretty generic field set. And I would imagine there's some customization clearly that you can do within it. And then essentially I would imagine they can get to a lower price point cloud-based.

Yeah. One stop shop. It could also potentially avoid an acquisition by my practice group.

Yeah. I mean, that's, that's, that's gonna be, that's going to be formidable.

And if you think about it the regional the local hospitals, I can now get the hospitalization data back to me as well. So not liking that interoperability word, but it does take care of that as well.

How about this, Microsoft closes $16 billion acquisition of Nuance. Can we talk about the nuance acquisition? What do you, what do you think of that one?

Well, it's kind of interesting.

I mean, now it's closed. It was announced and they went through the dance. And now that I guess they're their partners. Dance partners. Yeah, it'll be interesting. What do you think they need to do to be successful

Well it's only one part of the picture. It's not like Microsoft having enough assets to be healthcare, right. It's just one part of it. What's their intention. Yeah.

But I'm hearing a lot of success for Microsoft across the board. So they're not trying to get into healthcare per se. They are. They're I always call them the arms dealer. They, they provide tools. There's tons of tools, right? So Microsoft teams on the basic and office 365 on the basic, then you have you have Azure. We have two health systems two major health systems.

You have about three, three and a half billion dollar health systems that have moved to Epic and Azure for production. Correct. We knew that they were doing DR on it, but now we have full blown production in Azure. We have a bunch of data stacks going into Azure. People are saying it's easier. It's the same tool set. My team can run with it and and whatnot. So they're, they're getting some wins there. The Nuance tool set is an interesting one because that does get down more into the room and to the clinician,

Nuance is probably closer to mobile products as well.

Yeah, it will be fascinating to see what they what they do next.

It could be a a speech detects play and the analytics around it. 📍

📍 We'll get to our show in just a minute. As you've probably heard, we've launched a new show TownHall on our Community channel. This Week Health community. And it airs on Tuesdays and Thursdays. I'll be taking a back seat to some of these people who are on the front lines. TownHall is hosted by an array of talented healthcare leaders who are facing today's challenges head-on. We're going to hear from professionals and their networks on hot button issues, technical deep dives, and the tactical challenges that healthcare faces. We have some great hosts on this. We have Charles Boicey and Angelique Russell, Data Scientist, Craig richard v ille, Lee Milligan, Reid, Stephan, who are all CIOs. We have Jake Lancaster and Brett Oliver who are CMIOs and Matt Sickles, a Cybersecurity first responder. I'd love to have you listen to these episodes. You can subscribe on our Community channel. This Week Health Community, wherever you find and listen to podcasts. Now let's get to the show. 📍 📍

Talk about the AI players. I mean, we see AI in every booth. I know you're an AI skeptic in some, in some regards here. So if I'm going booth to booth, what am I looking for when I'm talking to people about AI and saying, all right, is this really AI or is it just marketing.

Even more important. Talk to me a little bit about your dataset. Tell me about the features that were selected. Tell me about the model build. What type of model is it? Ensemble is it. Nuero Network. What the heck is it? And then what are the outcomes? So clinically, I want to know what the data looked like, what features were selected. And I want to know everything from start to finish. The minute proprietary comes out.

It's time to hit the next booth. So you can't block lots of stuff. I think it's the message and what works great for a patient population in Orlando may not work as well in Newfoundland. It's going to have to be tuned for that, for that population. And I think that's something that gets lost out there on the floor. And that's the whole bring the clinical folks, bring the operational folks and bring the finance folks in because they need to understand it from data all the way to the results. And math isn't patentable. So stop.

I had a ... stop trying to patent it?

Yeah. Just stop.

I had interesting conversation with Michael Pfeffer with Stanford. Came from UCLA. The CIO. And he made that point. He made that point about the value of their local data and not only their local data, but the data from the ED between certain times it's different than the ED from other times. And there there's value to the global dataset. He goes, but not nearly as much value to Stanford for that, what they can do with their own dataset and really focused in on high quality data outcomes really manage the, the drift that you would get in an AI model and those kinds of things. He was, he was very much saying, you know what? Our, our dataset in Northern Cal is appropriate for our population.

It is. The demographic, the conditions and so forth. And the other side of that, that these folks, all decided you need several years worth of data. So those folks that are going on to Epic, going on to Cerner, switching out whatever, and they changed everything to PDF.

And now they only got six months worth of data. How do you tune a a data science product and that environment. You need that longitudinal record from that from that geo location.

With Clearsense in the world, are people really still like moving from one to the other and moving everything to PDFs? I mean, there's there's tools out there.

No, there, there are folks that are doing that. You absolutely, you better believe it. Which is you want to, if you're going to an archiving thing, you want to bring that data over and its natural state in its entirety because I'm telling you as we go forward, we don't know what data is and what features are going to lend them selfto have higher weights and higher values because we haven't done the complete exercise of running all the data that's obtainable through that through that life cycle. So there's a lot of work to be done.

All right. Let's get some conference stuff here. So I don't know how much time you've gotten to spend. What's the most impactful thing you've seen? Have you seen, have you walked boost yet or anything like that?

I've done a little bit of that. I really liked that folks from the Philippines or here. Folks from as, as all as the Israeli folks are here.

Yeah. The, the international players.

I love that stuff because it's all relevant. It's a completely different perspective. And there's some, there's some tech in there that is worth taking a look at.

So even though healthcare is different, I talked to some people from Saudi Arabia, some people from Abu Dhabi. And I asked them the different models. And to be honest with you, I mean, it's, it's different flavors of things we're doing in the United States.

It could be single payer, it could be a combination of some kind. It could be whatever, but at the end of the day, they're struggling with the same things and pulling the data together. In some areas they're more successful, smaller, small geography for smaller population, different government oversight and regulation. The HIE in one of those countries has a hundred percent participation rate. I mean, I don't, I'm not sure we have a state that has a hundred percent participation rate.

But the question is how much, how many times is it hit and how much retrieval comes it. In the states, we have the same thing, we have, we have some states that have excellent participation from a data ingress, but from an egress, how often are they hit?

The other thing I didn't, I didn't really contemplate is states here, we have, we have trouble going state to state, but it's not nearly the trouble they have going country to country. I mean, when you, when you, you lay these things next to each other, it's like this country is next to Turkey is next to this. Those countries have completely different

Intra. Interacountry operability, but you know, same thing here, this state thing. We could have a national HIE. There's no reason technically that we don't. Technically no. Technically, no, it's all the other stuff that surrounds it.

Yeah. It's, we've had some interesting conversations around around interoperability while we're here. I, I caught up with Micky Tripathi. I didn't have an interview with him. I just chatted with him for a little bit. Talked to Don Rucker, former well, actually for right before Mickey, he had Mickey's job and Interesting to see where interoperability is trying to go with the APIs.

But to me, I look at bulk FHIR and it's the thing that, we spend a lot of time talking about, about 21st century cures. We talk about the APIs for the individuals. The individuals are gonna be able to get their data and that kind of stuff. Right. And you hold a set of apps. But when we talk about bulk FHIR, there's a lot of application for bulk FHIR within my health system.

As a CIO. I see a lot of practical application for pulling large sets of data together, again for population health metrics for clinically integrated networks and those kind of things.

Yeah. So we do an acquisition, a large acquisition, or several of them. If we go that way, there's a lot less expense and it's a lot of time. Is it it's much more timely that we actually get people working together. So don't have to rip it, my EMR with this, he has a solution. It's something to think about. Long and expensive process.

I caught up with someone from Intermountain. I won't, I won't use the name and I asked them because they acquired SEL. SEL is on Epic. Yep. And they're on Cerner. I said, all right. So the deal is going to be final on April 1st. What's the thought? And he said, we're going with two EHRs. We're not, we're not going to rip and replace either of them.

Yeah. I think you'll see that more and more because Hey, what's happening in the last couple of years, things the money's pretty tight. And does it make sense to rip and replace in EMR if you have a technology such as bulk flyer and others?

And, and so one of the drivers as it usually is, was we, it would take us 30 days to close our book. Bring it, all that stuff together. And by, by going to a single EHR that that care complexity goes away. So I wonder if we're going to be able to, I hope we're going to be able to get away with it. Cause it makes no sense to do you know, these tens of millions of dollars, if not hundreds of million dollar projects to, to rip out a perfectly good EHR. Yeah, so it's.

All the service providers in this hall right now, we're not going to agree with that.

Most transformative technology that you've either heard of or seen while you're here?

I'm going to go back to the use of conversational agents, AI, and I don't like the term that whole digital front door, I'm not really cool with that, but yeah, just engaging, engaging patients initially and routing them properly. And then making that whole initial engagement clean.

Because there's not one front door is there. There's like, there's like a million front doors.

But look at we're, we're in the business of providing healthcare, right? We've done a horrible job with our call centers, scheduling all that kind of stuff. There's some really good technologies out there that are being used in the hospitality industry and logistics, and in other industries that we need to bring in and make them work.

Yeah, I agree. I've seen two computer vision things I think are interesting with AI models. NVIDIA on the backend. One is mammography and it's taking these, these mammo images in, and it's not doing diagnosis, but it's, it's 3D, it's processing through these NVIDIA chips. It's doing it in like 20 seconds and it's actually circling things saying. Look at this, look at this, look at this.

And that's exactly what it should be doing is assisting the clinician or even in pathology, Hey, take a look at this part of the slide or in your imaging just to, just as you describe it. But what we're setting ourselves up for is really bringing this technology into the home where we can do the diagnostics and a higher end monitoring in the home environment.

And it's basically edge computing, right? That goes back, it gets like we can go into the homeless, smaller devices and do that work, get it taken care of. And I think you're gonna see more and more of that, where the acuity of the patients is going to get greater and greater in the home environment.

Yeah. That's, that's interesting. Cause the gentleman, the CEO I was interviewing on that I always like to do this to entrepreneurs. I said, all right. So let's assume this company that you're doing over here, which is doing a great job and they've expanded you sell it, whatever. Push you out and you here to start over what we'll just start with.

And he said said dermatology directly to the patien. Same technology dermatology. He said, we could train those AI models to look for certain things and give information. It actually could be two way. Information right back to the patient and information to the dermatologist saying, Hey, you might want to make a phone call. Sure. Those kinds of things really.

Assisting. Always assist. So we look at it as intelligent assist. Not artificial intelligence. Clinically those that prescribe don't necessarily want to be prescribed to, but if there's technology that will assist in that process. Absolutely. Absolutely. Hey, take a look at this, this, and this.

Have you considered this, this and this over the last hour or so this, this, and this has occurred. You might want to look into something, but you know, this whole thing where you're gonna do the the dynamic. And this fantasy is not so much. We appreciate the help, but we don't want, and it's not a job security thing. And really honestly, when talking, talk about that, here's the, here's the real deal. Human beings can make mistakes. Right. And be forgiven. The machine gets to make a mistake once.

Yeah, it doesn't get to make a mistake. This is why we don't have autonomous driving yet. There you go. One death is too many for a for autonomous driving for a computer but how many people are going to die today.

Autonomous driving for adoption has to happen at once. You can't have another human being in a car with autonomous driving because they all work off of you.

Have you seen, have you binge-watched Upload yet? No. On Amazon. upload is essentially uploading your consciousness. Yeah. Yeah. And so but essentially in a world of complete autonomous driving, the guy dies by a car accident. And every time he tries to tell somebody that it was an accident, they just laugh at them. It's like, there, there are no accidents.

No, no, they are not. And Bill just to kind of take that a little bit further. There's a group in Stanford that is doing the following, which is extremely interesting. They are taking, and you know it up here, we have pedabytes of information stored. Pedabytes. Absolute pedabytes.

So you have more pedabytes than I do. I mean, what's your, what's your storage.

Yeah. But think about that. So what they are doing with this group is doing is transferring binary ones and zeros to DNA protein sequences, and then storing them.

Storing them where?

Storing them like you would normally see on a SSD drive or on a spinning disk. And getting down to the from a level of storage, a hundred to a thousand times the capacity. The retrieval is what's being worked on right now. So it can be retrieved in sub-second time.

Wow. That's that's amazing. Computer vision. The other thing is I just interviewed Artisight CEO up at Northern Western. And they're addressing the problem of staff shortages with simple cameras in the room, processing this stuff through AI. And there's, there's an old way of training AI models, where you essentially send all these images to fill in the blank with people in front of it. And they go, that's a coffee cup. That's not a coffee cup. And the problem is you take that coffee cup sitting on this table and. And the computer says, yeah, that's a coffee cup. Then you move it onto the top of Mount Everest and they go, I don't know what that is yet. So you really, I mean, you're constantly trying to train it.

But these new models that you're seeing at Meta Facebook and others, which is what's being applied here it essentially it's the computer training itself. You take out pieces that it, yeah, it teaches supervise. Yeah. Right, exactly. Unsupervised training. And so this tThis simple camera being processed again through this NVIDIA stack and technology is looking at this thing and it now knows the patient room. It's looking at that patient from going, okay. I know the patient from, what do you want to know? And you could tell it all right. Look for hand-washing. Okay. Look to see if the beds aren't done. Hey, look, to see if the patient needs to be moved.

Or even potential falls. Potential falls. or, How much actual interaction between the clinical staff and the patient.

The thing I liked about it. And your Typhoid Mary. Because you and I have talked about platforms and the power platforms. The thing I liked about it is he's putting it in the hands of these people and they're going, Hey, I've got another problem inventory of equipment and stuff. Can we put a camera here that looks at the inventory and when the inventory gets low, it'll kick off an automated process. Yeah. And he's like, yeah, absolutely. Hey, can we put this on our inventory, on our loading dock and whatever, can we put this in our. And it's like, yeah. Cause it's a platform that all it does is it learns the environment. It takes the thing with computer vision. And the, the whole idea of taking that data stream is we're trying to apply AI. And on top of all this bad data that we have, right.

That's an environment that is actually clean.

The image is clean. It's, it's something that we can actually act on. If the, if the computer can be taught to recognize certain things. The bins empty. Order more fill in the blank. Or a shipment just arrived on the dock, send somebody down instead of.

Sure. And if you got, if you got them in all the right places, you'll you'll understand that hoarding habits have already certain clinicians and certain units. Yeah.

Hoarding habits. Hoardinghabits. See, that's the benefit of, of you having worked where. I've Yeah. You've, you've been there.

We've got a whole bunch of you know IV pumps in a closet somewhere because we don't want to run out.

It's it's interesting to me too, because as we were talking, I had a CIO there, as we were talking about it, I said, talk to me about the case to the CFO. And she was like, well at the end of the day, every one of those things we just talked about, the answer usually is, Hey, can we go hire another person? We need a person to do inventory. We need a person that's.

Even since from a clinical perspective, somebody that's on 24 7 watch, you don't have to to a nursing attendant in that room. That takes care of that issue. And if somebody that's not supposed to be out of bed starts moving that direction. You can get somebody in there and take care of it. So you prevented a fall and then you prevented the dollars that are associated with that as well.

Fascinating conversation with some nurses at Atrium Health. Doing a virtual nursing program. Not replacing the nurse at the bedside. You always need that nurse at the bedside, but supporting that nurse at the bedside with virtual and you have a ratio that's pretty high cause a nurse at a virtual location can through cameras and other things react to the, to the nurses in the room.

Yeah. And you know why it's important to put AI into it because that nurse, myself being one of them, his attention span is only so much. So this is where, yes, the nurses present the nurses doing the surveillance, but you also have surveillance being done in the background that can then assist that nurse said, Hey, there's likely to be a problem here. This is where he really need to focus right now.

It was, it was interesting to hear them talk about the case for the virtual nurse is what you talked about. The workload and taking some of the workload off training. So you can have experienced nurse. Virtual nurse. And they said we're hiring all sorts of new nurses. Bringing them into the system. So you could have somebody who's less experienced, have oversight. Sure. In.

Mentorship. So, so the nurse has already got a medication that she's never given before. And you've got some awareness of that. And then that mentoring can take place.

Yeah, that was it. I mean, as she was talking, I was like, that's it, it was really creative. And then the other thing is you could have that nurse in the back. You get busy as a nurse, you can just say camera's there, microphone's there and say, Hey, I'm I'm right now getting ready to do an IV or whatever. Can you document and that person who's back there monitoring several rooms can do that with documentation. So you're taking away the context with all the things that get us to be distracted and whatnot from the nurse. Interesting interviews while I'm here. The pandemic caused us to be very creative. It changed the environment so much that we worked. We weren't as tied to the static. And I think that the creativity is out there. Still some problems in healthcare though.

There always will be. And there always will be. That's why we're here.

That's that's. Yeah, exactly. I mean, I, I think we're seeing some some interesting things. I'm surprised that hall is as big as it is with as many solutions it is, but I'm not sure why I'm surprised. I guess I'm surprised because I know.

McKesson is still around.

Yeah. McKesson is still around. Yeah. That's interesting.

McKesson was a big one. Back then it was all health information management systems. Like it was all we call it medical records at the time. Right. And then there was some financial ethos. So the original introduction of technologies has dropped the bill.

I'll tell you there was a, I think there was a good 300 nurses in that room yesterday for the, like the informatics symposium. Yeah, it was good. It was a good dialogue. They also have their case studies. I think they put around the room. Yep. The posters. Yeah. I took a picture of all of them. They were, they were so good. Excellent. I liked the work that's going on in nursing. Nursing is in my conversation, the rate at the bedside.

Bill, bill it's interesting that you so I'm an advocate for the nurses. I've been teaching it. I've been preaching it. Who better? So if you're talking about data science, you've got domain knowledge, you've got statistics and you've got technical skills, programming language. I can teach a nurse programming language she's already got, or he's already got the statistics. What they have is a 80% domain knowledge. Now taking a data scientist from somewhere else and dropping them into healthcare. That's hell. That takes a long time for them to get the domain knowledge. So over the last couple of years, that's what I've been doing is turning nurses into clinician data scientists, and they've been working out really well.

Is that, are you talking in your company, are you talking at Stony Brook?

I'm talking stonybrook. I'm talking just advocacy and getting nurses interested in data science. I mean, who implemented the EMR? Right. The EMR was on the backs of nursing. Nursing Informaticists. And then we got into optimization. We have a huge opportunity now from a data science perspective to actually take that next wave and actually make it ours. So who best than nurses to take us into the machine learning artificial intelligence world.

Yeah. And you're living proof of that.

Yeah. Or as I call it intelligent assessed.

Yeah, absolutely. Charles. Thanks for your time. Appreciate 📍 it.

What a great discussion. If you know someone that might benefit from our channel, from these kinds of discussions, please forward them a note, perhaps your team, your staff. I know if I were a CIO today, I would have every one of my team members listening to show just like this one. It's conference level value every week. They can subscribe on our website thisweekhealth.com. They can also subscribe wherever they listen to podcasts. Apple, Google, Overcast. You get the picture. We are everywhere. Go ahead. Subscribe today. We want to thank our news day sponsors who are investing in our mission to develop the next generation of health leaders. Those are CrowdStrike, Proofpoint, 📍 Clearsense, MEDITECH, Cedars-Sinai Accelerator, Talkdesk and DrFirst. Thanks for listening. That's all for now. 📍