Is Digital Pathology Evolution Enhancing Histology Quality and Pathology Data Usage
[00:00:00] From buying a whole slide scanner, even before Leica acquired Aperio, through founding a digital pathology technology company to support rural hospitals and founding a digital pathology practice where remote pathologists actually could support those hospitals with the use of the technology that was invented in the technology company to a digital diagnostic foundation.
Learn about the newest digital pathology trends in science and industry. Meet the most interesting people in the niche and gain insights relevant to your own projects. Here is where pathology meets computer science. You are listening to the digital pathology podcast with your host, Dr. Alexandra Zhurav.
My today’s guest is a personification of another digital pathology bridging journey. I’m super excited to present him and his story to you today. Welcome digital pathology trailblazers to another episode of the [00:01:00] podcast. Today, my guest is Dr. Matt Levitt. He is a pathologist. He’s also the executive director of the Digital Diagnostics Foundation.
This does not have pathology in the name, but it very much is his combination of his digital pathology road. He is the founder and was the c e O of Lumea that is a thriving company still in the digital pathology he founded and was the c e O of Path. Uh, so welcome to the podcast, Matt. And how are you today?
I, I’m great. It’s my, Honor and pleasure to speak with you, Alex. Thank you for having me. Matt, tell the listeners about yourself and about your journey, about your digital pathology journey. Well, okay. So you’re right. Um, even though it sounds like I’m doing all these different things, it’s all one vision.
And it’s just, I started off as a pathologist, seeing the potential for digital pathology and we just wanted to go out and practice digital pathology. That was really all I was wanting to do. And then, uh, in 20. 13 went out and bought a [00:02:00] scanner. I was sort of excited and a little, which one did you buy? Oh yeah, I, I bought the, uh, it doesn’t exist anymore.
It was the Leica SCN 400. It was before a Perio and it was, it was enormous. It was huge, but there was a good scanner for the time. And, uh, But it was like taking out a second mortgage. And at the time I was at Intermountain Healthcare in Utah and doing well enough financially that I was maybe a little overconfident and convinced my wife and my partner did the same with his wife.
I, we said, Hey, we’re going to buy a scanner and we’re going to open this digital practice. And then we did what I think a lot of people that bought scanners at that time did, which is when the scanner came, we were really excited. We started the validations and then we realized this is completely impractical because it was just, it took longer to scan the slides and look at them on the computer than it did just to, just to look at them in the scope.
And we would end up mailing [00:03:00] slides, uh, that we thought we were going to scan just because it was easier. And so we started asking ourselves, why is digital pathology not practical? And it very quickly came. I guess there are still people asking themselves that questions because maybe the like scanning time and the viewing and all this stuff is already ironed out, but there are a lot of workflow hiccups that still need smoothing.
I would say. Yeah. And I think everybody. Broadly in the community, I should say there is this question of return on investment. I mean, as the, as the, as the profit margins, the margins get thinner and thinner for pathology because reimbursements are coming down at the same time that this extensive investment is required to go digital.
Uh, people are asking themselves, what is a return on investment? And that’s exactly where we were. We’re like, wow, we just took out a second mortgage on this, on this instrument. How do we make it useful? And so it’s maybe a little different when it’s your own money that you put out there versus your institution that has to ask that question.[00:04:00]
This was keeping us up at night. And so we started talking about it on our lunch breaks and saying, well, okay, what if. We were to think about the pre analytic workflow for digital pathology, would we change it? And there’s so many things about the pre analytics and pathology that need to be revised anyway, that need to be changed anyway, because if you’re engaged in molecular diagnostics or in immunohistochemistry, you realize That the variability in pre analytics really is not good for, you know, for these two revolutions in diagnostic medicine.
One is digital transformation and the other is molecular diagnostics. In both of them, there needs to be more effort put toward pre analytic standardization. So with that in mind, we thought, okay, well, what if we were to re standardize or think of a new way for handling specimens? That would fit in with cloud computing that would fit in with better, uh, tissue preservation that we’re talking about.
Cloud computing in 2013. Hey, [00:05:00] 2013, the cloud was big. It wasn’t AI back then it was, and so we were thinking, what if we were to, what if we were, you know, Well, let me, let me go back our practice covered a vast geography in Utah. So we were, we were getting specimens from the, uh, reservations in Arizona. And, and so, you know, we covered most of the state of Utah specimens were coming to our hospital and the disconnect between what was happening in those clinics.
The only thing that connected us as pathologists to those clinics was a requisition sheet and, you know, a, the occasional. Visit that we would go as medical directors to visit those those labs, but it really on a day to day basis. We were really disconnected. And so we thought, okay, how do we solve all these problems?
We could do that with new technologies, but we have to rethink some things that are really deeply ingrained in pathology and go right back to where the patient is. And so we, we came up with this concept of biopsy link biopsy link is [00:06:00] a digital bridge, a system that bridges what’s happening at the bedside with for the patient to tracking the specimen as it goes from the patient.
To the laboratory, and then and then that’s not where the specimen stops. It doesn’t stop with the slide. You end up ordering ancillary tests. And so parts of that specimen end up getting sent all over. So how do you manage all of that? In a way that’s more efficient. And so that type of thinking, uh, sort of rethinking the whole thing from the ground up made us think this is a whole company and we’re not, we’re not equipped to actually do that.
So we got with some engineers and said, okay, these are the things that we’d like to accomplish. Can you create, help us create a system that accomplishes these things. And so what started off as wanting to be just practice digital pathology turned into Lumea. Which is, uh, the company you introduced, uh, and so I ran that until it was no longer wise to have a doctor running the company, which I learned pretty quick.
I, you know, I, I didn’t want to be [00:07:00] in business. I don’t have to insert a question here because as you’re saying this, this kind of answers my thoughts. I was like, okay, uh, you as a pathologist who just wanted to practice digital pathology. Suddenly like started looking for solutions that grew into a business that you decided, okay, I’m going to run this business.
Like how did you want it in the first place? Do you have any background? Do you have like your character traits? Like how, how did this happen? And how did this happen multiple times? And then we’re going to explain what Lumia is as such, at least like, um, you know, a high level, because I’m going to meet with them and have them on the podcast as well, so we don’t need to go into detail, but I want to set up the stage, what it is, but first about you, like how you are still a doctor, right?
Now you’re a doctor. Yeah. And up until a couple of years ago, I was practicing and I still want to practice. So I just. I got so engaged in the, in the business that I didn’t feel [00:08:00] like I was doing anyone any service by, by practicing because I was not giving to it what I want to. But, but here’s, here’s how this happened.
I, again, I didn’t want to, I specifically chose not to do business because, but you can’t get away from, I guess, your genetics because all of my family are business people. My, my dad started a business. My brothers start businesses. I’m the youngest of six sons. Everybody’s sort of in business. And, and I was like, I don’t want anything to do with that.
I want to just purely practice medicine. I don’t want, I don’t want to do that. Then I got so excited about the innovations and then realized this doesn’t go very far without money. It’s not sustainable. Without making it into a business, right? Without making it into a business. So I found myself trying to Drive change in medicine through this vehicle of, of creating a sustainable business model and I understood the market and what would appeal to and what the needs are in the lab [00:09:00] medicine market, at least the anatomic pathology ecosystem.
But really didn’t understand what it means to, you know, to, to run an efficient business, but I knew enough to get it started. And I had, again, I, I didn’t lack in confidence. If I would have known what I was getting myself into, I would have thought twice. It’s kind of like, it’s kind of like having kids. If you knew all the things that you knew, there are days.
That you’re like, wow, would I have made the same decision? You you you love them, but there are days when it’s really hard and and yesterday for me. Yeah My both kids at home and I was trying to like semi take care of them and work and none of these two things work Yeah, in the end, you would never go back on that decision, but the reality is when you start down the path like that, like, like when you first have kids, you have no idea what you’re in for.
And that’s, that was true for starting a business. I, I jumped in with confidence, excited, and I still am excited about what [00:10:00] Lumina is doing, but I had no idea what it really meant to run a business. And very quickly, I could see there are people that can do this a lot better than me. See, this is a cool thing because there are a lot of good business people that can later run businesses that are already set up to support pathology.
But if you start the other way around, you don’t have so many pathologists that could do the heavy lifting to like straighten it and bring it to a place where it actually Delivers value in the places
where it’s supposed to deliver value. So you like started the right way and then you left and found bear theme. Yeah. It is the reverse of what, the way it usually happens. And, and there are good, good things and bad things about that. But I very quickly realized I was in over my head on running a business.
And I, it’s not really, again, it’s not where my interest was. I wanted to be in the lab in the early days of Lumia. We didn’t go out and get massive funding. I took some big personal financial [00:11:00] risks and my family took some financial risks to support the start of this, but the company was actually sustaining itself by having a CLIA certified lab.
In house, so that was important. Not only from more important than the financial revenue that generated. It was the fact that we were learning how to practice digitally and we were learning how to not only the technologies, but how does a digital practice work? There’s 2 levels of innovation here. And I learned over time.
They’re different at the beginning. Digital practice and technology innovation were kind of the same thing. That was all in 1 company over time. As the practice started to grow and there were all kinds of questions about regulatory and where does this fit? Uh, we realized. These are two different types of companies running a clinical services practice that has to deal with CLIA and CAP and inspections versus a technology company that needs to work with FDA and, you know, [00:12:00] it gets really fuzzy if you’re all together.
And so we had to make a really difficult decision. We made the decision actually in about 2018 that we needed to split them apart. But it’s really hard to split the lab and the technology technology company. So 1st, it was all together. And that was how was surviving was by doing clinical services and essentially paying their pathologist and other pathologists less than what they were earning, but supporting this, uh, Innovation addiction that we had.
So, uh, so, but at the point that it really started to take off as the technology ripened and as we started to gain traction in the market, we realized, Hey, these are two separate businesses and they need to be separate because who, you know, the other problem with Lumair was who wants to buy a technology from a company who’s competing in the marketplace, you know, for, you know, and, and so we just realized that needs to happen, but it was hard because.
It meant that, oh, so you mean [00:13:00] which lab would like a technology from a different lab that’s doing the same services, a lab who’s competing with them for sure. So, um, the healthcare can be very territorial. I don’t know if you feel it as much in the veterinary world, but in the healthcare systems, if you’re in a, in a city like Boston or New York, where you’ve got these competing health systems.
It’s brutally competitive and so, you know, it’s not likely that one health system is going to buy technology from another health system when they’re competing head to head and in that marketplace. So, you know, for instance, if I’m not going to mention any names here, because I don’t want to get your.
Podcasting trouble. So I’ll stay away from that. But, um, yeah, but although if you don’t get me in trouble, please do mention any names you can mention. Cause it’s interesting. So if anything, then you’re in trouble, right? Yeah, that’s right. I, I, yeah, I’ll just, I’ll stay, I’ll steer clear of anything that stirs politics for either of us, but no, no politics, but the mentioned names, if you [00:14:00] can, of companies situations.
Yeah, because I do want to paint a picture of the landscape and, you know, these are all players in the landscape. So if stuff is out there and you can manage, you can mention, then go ahead. Yeah, well, I’m going to, I’m going to mention two that would be obvious to anyone just to illustrate. So if, if Quest diagnostics who competes fiercely with LabCorp, if Quest has a new technology, it’s unlikely that LabCorp is going to buy that technology.
LabCorp is going to say, well, we’re going to develop our own to compete with Quest. They’re not going to support one another. And anatomic pathology on a smaller scale, if Lumea was a lab and it has this technology, it’s unlikely that LabCorp is going to buy it if they’re competing with one another. So we decided, okay, we need to split these apart so that PathMet.
Can innovate in the digital practice of pathology, Lumea will innovate in the digital technology that supports that. So the new company was [00:15:00] PathNet and this was the digital practice. Whereas Lumea, the technology got separated to be able to serve other practices as well. Exactly. Exactly. So now Lumea can sell its technology to PathNet’s competitors.
So let’s, let’s take a step back because I want to hear the definition from you of a digital pathology practice. What is a digital practice? And I guess a PathNet is going to be the example. Um, and then I want to know what technology did Lumea support the practice with? Great question. So what, one of the things that we learned.
Very quickly, and I guess 1 of the things that led to this is that when you add digitalization to a histology workflow, you increase the cost substantially. So, by increasing the cost, that means that you have to have a very high volume going through that you need high throughput in order to make that. a viable return on investment.
Right now in the United States, histology [00:16:00] is happening in every small hospital in lots of physician clinics. Histology used to be reimbursed very well. And so a lot of clinics and hospitals got into the business of histology before the reimbursements got cut. But there’s, but that’s still happening where a lot of, there’s a lot of small boutique histology labs all over the country.
And what we realized. It’s in the sort of the business of digital practice. It makes no sense for these histology labs, they can’t, they can’t buy each by scanners and each by a digital system. And why would you, if it meant that now your pathologist is going to come and sit in your hospital and just look at it on a screen there, I mean, there’s no advantage really over Right.
Yeah. over a microscope. What’s needed is to be able to say, okay, we’re going to centralize slide production and digital image acquisition, which enables us to be able to create the volume, you know, to, to have the [00:17:00] volume and the throughput that supports that technology and then more efficiently delivers that, uh, to lots of different clinics and hospitals.
So to say that more succinctly, what we could see is that digital transformation is To consolidation of histology, there are going to be fewer histology labs, but they’re going to be bigger and histology labs are going to need to service many different institutions in the United States that creates problems that could create problems.
For hospitals that rely on that revenue stream. And, and so it, and that’s an impediment to digital transformation. It’s like, Hey, I still need the revenue stream, but I lose profitability if I go digital. So that’s, that became an impediment to people making any, you know, the decision to go digital. It made no economic sense.
In the PathMet model, we created what are called digital histology centers. And in the digital histology centers, uh, [00:18:00] those serve many different. And the digital histology centers are not billing centers. They don’t, it doesn’t actually have a, you know, it’s CAP certified CLIA certified, but it doesn’t actually, it’s not the, um, the billing entity that enables it to be able to serve, you know, 10, 20 different laboratories, consolidate that volume, those laboratories.
Don’t lose the revenue of histology because this is an important nuance. The technical component billing is based on where the gross examination happens, not where the slides are made. So what’s important is that the gross examination, the accessioning gross exam that needs to be done by somebody who’s qualified, that needs to happen on site based on the regulations.
But they didn’t have to have histology anymore. But they don’t have to have it. To get paid. To get paid, yes. So, so, so by doing that, you’re able to, so they maintain the billing, the, the [00:19:00] hospitals and the clinics, but now they don’t have to support these boutique histology labs. And most of the time, the small ones, the quality is, is not that is not very good because if it’s small, it’s usually made up of people who are moonlighting at five, you know, three different labs.
And they, and they just can’t give the time and attention, full time and attention to it. I am with this podcast. I’m learning like so many work arounds the, the, the US healthcare system. Like how can, in another podcast, I had a guest who was telling me how you can increase the amount of money you get by having a different level of consult when a dermatologist is showing a digital slide to the patient, even though the diagnosis was made somewhere else, but.
by a dermatopathologist, but like, there is a way to capture the money for this. Yeah, it’s, it’s really, uh, people have gotten creative, right? That’s what it takes. I suppose the thing that I like about this, it’s not just a way to [00:20:00] capture revenue, it solves what I think is a quality problem. the quality problem.
It, like I suggested, is the problem sort of started when the reimbursement for histology was so good that people started making histology by itself, a business and putting it, there were, you know, there were companies that went out there and their business model was install histology, labs everywhere.
Mm-hmm. and, and, and that. It’s generally just not, it’s generally, I think, difficult to have high quality if you’re replicating it everywhere and it’s incompatible. That model of having histology labs everywhere is incompatible. So not only are we trying to, you know, make it. economically viable for digital, I think we’re actually increasing the quality, improving the quality of the histology in those places.
And I’ll give you an example. One of the first pathologists to join our ecosystem was Dr. Adam Cole. Dr. Adam Cole was based in Arkansas and his Practice was he was working for another company and they had set up [00:21:00] histology labs in many different physician owned labs and he would take riding his motor home from one from one practice to another and, and, and, and just go, go, go in a, in a, in a circuit and, and who was like taking care of all the slides in those clinics.
I mean, it was just, it wasn’t great quality and wonderful. This is hilarious. It was not a good quality of life either. I mean, it’s, it’s, uh, so you should probably put him on the show and let him on
LinkedIn, but I want to hear the story. Yeah. Anyway, uh, how did the pathology increased his personal life quality? Because actually does it for me because I go to Poland, um, once or twice a week. And if I wanted to take vacation, I would probably like. Would be able to stop working and because it wouldn’t pay off, but I can just work with digital pathology.
So I hear you. That’s I, that’s what I’ve been, [00:22:00] I’ve been dealing with that now. It’s tough to take a real vacation. Yeah. Yeah. But it does enable you that freedom. Uh, you can do workation. Yeah, that’s right. But, uh, so Adam very quickly convinced his clinics to centralize their histology and that he was able to practice digitally from home and you know, he would, uh, and that, and that was when that started in probably about 2016, 2017.
So let’s talk about this, this very thing that like, if. If COVID didn’t happen. Exactly. So I was wondering why is digital pathology not widely adopted in a CLIA setting where you basically can take it, validate, do whatever, right? Why it’s not there? Why did we need to wait for the FDA clearance of the Philips scanner in 2017 as Such a big milestone.
Why did we need a pandemic to like hear that? [00:23:00] Oh, finally you can work from home. Like logically you could have done that before already if you wanted to in a compliant manner, which is what you guys did. But tell me a little bit about this dynamic and this like transition, maybe cultural or societal transition.
It was a little bold and admittedly, it’s probably not something that if you were Large lab. I don’t know that they, they would have done this, but, but we took the approach. This is a lab developed test. It’s a new technology that doesn’t necessarily fit nicely with the regulations. And yet the improvement in quality from the motor home.
pathologist who was compliant, but, but the quality, this is the best he was compliant. But, but it was, but it was a mess and versus what we were doing, which tracked the specimen. I mean, I mean, there’s just the level of care was so obviously better, but we also [00:24:00] recognized that it didn’t fit perfectly with.
and CMS and, you know, I took advantage of the fact that I, my brother was the former secretary of health and so that’s a good advantage to have and I just, I asked, can you make an introduction to the head of CLIA and he just sent an email. I mean, he didn’t, he, there was no, obviously he wasn’t the secretary anymore, there was no, it was just, I needed to be able to say, Hey, we’re doing something here that.
We recognize it doesn’t necessarily fit, but let us show you why this is. This is superior to what’s happening out there. And they were all ears. And they were, they were, they were like, we, we, we recognize that digital, they mean Clia, the director of Clia and her deputies recognized that the regulations don’t necessarily fit well with a cloud based.
System and that the world was going in that direction and they essentially said help us figure this out And you [00:25:00] know, please send us a white paper that we can distribute to the different inspectors the different You know in the regions and to the different max max or the or the contracted payers for for CMS So that they’re aware of this and that’s what we did and so that that probably gave us a little more confidence To venture out in, in something that wasn’t quite there, you know, what we had going for us was we had their assurance that they knew what that we are being transparent.
We had the proof and documentation that our lab developed test method was superior. To what was happening before and so if ever it was questioned from a point of view of this isn’t in compliance We would just say we’re looking out for the patients We’re doing something that we can in a documented way show that this is better for the patients We’ve talked to the head of cleo.
We’ve talked to everybody that is that will be inspecting us What else do you want us to do? Are you really going to hold back [00:26:00] patient care and progress? And so with that? That mentality, we sort of moved into it before it was cleared, but those regulations are still fuzzy and there’s still a lot of ambiguity and there’s still work that needs to be done to update our regulatory framework so that it fits it.
With digital diagnostics and digital diagnostics is different in that there’s a lot of things happening that are important for the patient that aren’t necessarily happening physically inside the lab. They may be happening in the cloud and maybe happening in a system that doesn’t belong to and isn’t in full control of in the lab.
So there’s lots of things that, you know, that, that we are all are trying to grapple with. In this new sort of digital diagnostics environment, um, and that, and that is, that was sort of the beginnings of that journey when we helped this pathologist get out of his motor home and start [00:27:00] practicing on a computer.
So just briefly, what did Lumia enable the hospitals and the PathNet to do? What is Lumia? What technology is it? It’s not scanning. No, it’s not. It’s not. It’s not scanning. And when you talk to Lumea, let’s see, I’ll pull out some of these things when you talk to Lumea. I mean, I could share with you some things, some videos or whatever, but it’s probably better to have Lumea do that.
I’ll just describe. I’m going to have them. I already, they already scheduled the call. So all good. I just want you to tell me from a practicing pathologist, what was the system that you were using? What was this doing? So, so first of all, it enabled, uh, rural hospitals or hospitals that don’t have a histology lab to connect with clinics and their clinicians in their, in their surroundings.
And the way that we did that is we actually went to the clinics where biopsies were happening. Uh, we started in urology, uh, but it has application [00:28:00] everywhere. And we went to the urologist with something new. So I don’t know if you’ve seen this, this replaces six formalin jars. So in process, I saw this in their, in their presentation, it has, it places the biopsies, like several biopsies in the row with like some gels in between.
Yeah. So, so there’s this, and then this is the board biopsy board. And so instead of, uh, soaking in formalin and fishing out of. It’s placed on a special sponge that is soaked in formalin and it permeates the tissue very quickly so that it fixes the tissue quickly in this. And so 6 biopsies are sent in this rather than having to label 6 jars.
It’s gimmicky. I mean, you know, the idea was we need to be able to capture market share and you start in the clinic. And so if you go to the clinic and say, hey, send it to us because we’ve got this cool little. device. Yeah, but then you can cut them all at the same time and and they are all aligned. I saw it.
So yeah, [00:29:00] yeah. Well, that’s that’s an interesting part of the journey. So this biopsy board was a way to. To essentially make it give the urologist in this case, a reason to send to a base lab rather than to any other lab. It’s it’s it’s it’s it’s a value proposition. Hey, it’s easier. You’d have to label us and.
It turns out it preserves the tissue better, you know, because it, it, it fixes flat rather than, uh, squirrelly, you don’t have to like stretch. So I’d like to say that we were really smart and, and that we designed it with, with all that in mind. The reality is it was the reverse. Um, we, we found that biopsy chip that you referred to and that’s, that’s the, the gel or the, it’s, it’s an artificial tissue matrix.
It was developed by a brilliant scientist pathologist in Romania. Oh, really? Go Eastern Europe, Eastern Europe. So Lumea is actually a [00:30:00] Romanian word. Um, part of Lumea is still in Romania. So you go to Poland. I go to Romania. Let me know when you’re in Romania, then we can just like, you should come and see, and see that in Romania.
That’s so cool. Yeah. So we actually, what we realized early on was. That in order to make digital pathology valuable, we needed to come up with a better standardization in the way things were presented to the image viewer. And so it just so happens that at the same time we were thinking about that, it was about 2015, we came across a YouTube video from Romania, uh, this, sorry, new shot, uh, that, that was, was showing this, this biopsy chip.
And I just thought, wow, here’s to the power of social media for digital pathology. Whoever is on YouTube, click thumbs up and, you know, leave a comment. [00:31:00] It was, there, there is a, especially it’s one of the oldest interest groups on the internet. It’s the histology interest group, uh, histo net, uh, way back.
And so he puts, he posted something on. HistoNet a link to a YouTube video. The YouTube video had about 30 views. It was like, you know, it was it was not Oh my goodness. This is so cool But uh, but when I saw it one of our histo techs who’s part of that interest group said hey, Dr Levy, you should see this really cool thing and I saw it and I thought wow that would if you combine that with digital You can standardize And this is going to, so I also, I lived in Romania right after, right after the big changes in 91.
And so I speak Romanian. And so when I saw that, so 30, 30, 30 views on this video. And I thought, but, oh, but he’s a Romanian. So I I’m going to call him. And so I called him and because I was excited, you know, I didn’t have a lot of chances to [00:32:00] speak Romanian anymore. So I call them. So where are you in the army?
Where, where, why were you in Romania? Oh, I, I was a, I was a missionary for my church right out of high school. In the guys with the, the badges that I was one of those guys as a, as a 19 year old in Romania. It was a crazy. Yeah, yeah, you never know where this podcast is going to go,
but it gave me the, I think the courage to call this guy, uh, in Romania and we hit it off and I jumped on a plane. I went back to Romania and we started a collaboration and then eventually. We merged his company with our company. And now we’re, now we’re all Lumea and we adopted the Romanian name, uh, Lumea.
So, which means, um, sort of people of the world, which is, I think anyway, but, uh, that’s, that’s how, uh, the chip got started. And, and so it was the chip first, but the problem was you had [00:33:00] six specimens. In the slide, but we needed, and I guess what we, what we’d hoped to try to do is get the clinicians, the urologist to put it straight into the chip so that we never have to touch it.
But we, after 2 years of trying, they didn’t. So, so this was just the training wheels essentially for the, for the chip, but it turned out that the urologist loved it and it led to a whole new sort of system for lots of biopsies. This is for GI and small biopsies. Where, where, um, they place it directly onto these, uh, special sponges and it fixes quickly, but we’re able to image and document the specimen right in the cassette.
And so a human never has to touch it again. Uh, it’s a very So you basically match the, the picture with the slide and then you know which one is which. That is like so cool. I mean, it’s so simple, but like, yeah, why don’t we see it all over the place? Right. It’s amazing. We’ve [00:34:00] just got stuck. We’ve been stuck in this way of doing things for so long and I’ll say it really helps to have non.
Physicians thinking about these things, because before I started working with the engineers and students and things, you know, I was sort of stuck in my way of thinking, but if you start asking people that aren’t stuck in the same paradigms. They’ll come up with completely different ways of doing it, and sometimes those things are actually better, and I think that’s one of the things that we tried to really encourage at Lumea was, they call it DaVincian innovation is the term they coined, which is Uh, just maintain this sort of childlike curiosity and a willingness to look at things from a completely different point of view.
And everybody in the company may have a good idea. It’s not just the doctors or the senior engineers. Anyone can. And we’ve seen that, you know, this was actually the idea of a software developer. You know, I wish it was mine, but [00:35:00] you know.
But, but I think if you can foster that kind of culture in a company, it, it really, it’s amazing what, what comes out of it. Yeah. To be able to foster this kind of culture, you’ll have to like be able to say, like you just said, I wish it was mine. It’s not helpful. So let’s do it. Right. Because I think maybe it’s part of human nature that your ego just blocks things that don’t come from you that have such a great impact.
I think it’s something we have to like, through in general, like, I wish I was as smart as all my guests, but I’m fortunate enough to have them on the podcast. So that’s like my contribution to the, the advancement of digital pathology. So yeah, super cool. Okay. So you have the chip. We’re wandering, we’re wandering through this story, but that’s okay.
That’s kind of how the story is. So I guess, so the chip, uh, sort of. Open this idea of [00:36:00] creating standardized embedding and the board is the idea of okay, different type of tissue transport that links, you know, that is standardized and as a reason for the clinics to adopt and you had asked the question, what do we do for the hospitals?
Well, now, all of a sudden we are capturing the. the tissue around a hospital and our pathologists are then on staff at the hospital and, and yet the hospital doesn’t have to bear the costs of the histology. All they have to do is link into the digital platform. And so without having… So you didn’t have the pathologists anymore.
They were at the hospital. Well, that’s what started to happen in PathNet. We started to work with the local pathologists. And sometimes if there was not a pathologist on staff at the hospital. We would find a pathologist in path net to say, Hey, do you want to be, do you want to join the, this hospital?
They need a pathologist, uh, to be able to, to be able to have specimens come through that [00:37:00] hospital. And so, you know, one example is we have a, a rural hospital on a Native American reservation in the mountains in Utah, middle of nowhere. But the pathologist on staff there lives in Long Island in New York.
Uh, and so this, this rural hospital has like this world class pathologist on staff and he signs out from Long Island, but you know, it really opens up. Lots of doors, both on the histology side, a small hospital doesn’t have to have a, you know, doesn’t have to invest in the scanner doesn’t have to invest in the histology and can have a pathologist who doesn’t necessarily live there unless they, if they need somebody for frozen sections, then they want somebody actually, we’ve got.
We’ve got pathologists doing cow pathology. You do? Yeah. Yeah. We’ve got Path has pathologists doing frozen sections in Colorado who live two states away. And, uh, you know, you need a, you need a good tech there, but it, it really has, uh, enabled these rural hospitals to have access to top tier [00:38:00] pathologists and it gives those pathologists the ability to practice in lots of places and yet have standardized histology, you know, you know, high quality.
They don’t have, they can practice in lots of places without having to have a motor home. So imagine that imagine. Yeah. Yeah. So well, that’s what it is doing for those hospitals. I think what we’ve done so far has been focused on community practice, rural practice. We have path net i’ll i’ll speak to path net because I, you know, the man now has lots of different customers But, uh, you know, PathNet now has…
So Lumea was first, um, you found it, you left to PathNet. Oh, I need, I need to tell that story and then I’ll just… Yeah, yeah, yeah. Tell the story about that. So, so I mentioned that in 2018, we realized… Your editor’s going to have to take this and reorder it. No, we’ll leave it as is. That’s okay. Okay, we both take.
My guest, my Trey Blazers will need to listen to the end. Okay, [00:39:00] okay. Hopefully they’re still with us at the end. But we realized in 2018, we had to split Lumay and PathNet and it took the pandemic to really give us the motivation to say, okay, we’re going to do this. And it was the, the demand for a digital pathologist in the pandemic and that really drove that.
So we spun off PathNet and because I was more, I’m more interested in the digital practice, you know, I love the innovation, but if I had to choose, I I’m really. I really love the practice. So I, I went with, I went with PathMet and wanted to focus on getting this digital practice off the ground. And so we had three digital histology centers, DHCs, one in Michigan, one in Arkansas, and one in Utah, the original one in Utah.
And through those three digital hubs, We’ve centralized, uh, tissue and histology from over 200 clinics, six hospitals in 32 states. And so that’s, [00:40:00] and that’s, that’s just PathMet. How big is PathMet now? It’s, it’s probably got about 40 employees. I think maybe five or six, five, maybe full time pathologists and a bunch of contractors who are in those hospitals.
But, you know, it’s, it’s a digital practice really, but it’s networked through these relationships with, you know, we’re networked into probably hundreds of different small labs. You know, the ones that used to have their own histology now are outsourcing their histology and contract with these pathologists.
The interesting part about that was. While I feel great about the quality of the service that’s happening, you know, we’ve improved that and, and I think that that’s without question. One of the interesting parts is, okay, the labs, each of these small hospitals or small clinics, the data still, the data and the tissue still belong to them, even though they’re being serviced by these contractors.
And in order for us to really harness the [00:41:00] value. of all of this standardized, aggregated data. It needs to be managed in a way that is ethically sound, where there’s, you know, ultimately, tissue and the data belong to the patient. And it’s a challenge, you know, large academic centers, they have, uh, staff, and they’ve invested heavily in research infrastructure that enables them to utilize the data being generated in those labs.
In those large centers, the problem is that’s a very small swath of the human population. The academic centers tend not to get every socioeconomic group. Not every ethnic group, uh, makes it to those academic centers, which is a big question. Question you’re saying that the academic centers. already took care of the legal aspects and of the let’s call it logistical legal aspects of using the data from patients for research that this is like being [00:42:00] taken care of at what and you know what i’m getting at because i i need to mention this because there was a wave a little bit of a let’s let’s call it i don’t know ruffled public opinion.
Um, the moment the first AI diagnostic support tool was released, uh, by page AI, the prostate diagnostic support tool. And there was a discussion that I followed on social media that, Hey, all this data, they came from the center, but it was actually generated. It came from patients. It came from the pathologist who would provide those diagnosis and like, what’s the framework of just taking it and making a medical device out of this?
And that’s my question. Like, what’s the framework? Who has it? Who doesn’t have it? And that’s going to take us to your current appointment. Great, great question. And I’m glad that you brought that up. [00:43:00] Uh, and. And so I would say, have the academic centers figured this out? No, they haven’t invested in trying.
The problem is they are attacking the problem of research in the same way they’ve always looked at it with clinical trials. And I think digital diagnostics is fundamentally different and there are, you know, it’s not all staying inside the health system and there are other parties with. Different interests that are going to be, you know, using that data, building things out of that data.
And there’s all kinds, there’s a whole slew of ethical, social concerns around that, which is a whole nother discussion. We might need to have at some other time. Yeah, I’d love to. And there are people that know a lot more about that than I do, but I can see the problem. I don’t, I don’t know that, that I have all the answers for it, but, uh, running a digital technology company, [00:44:00] Lumea, gives me the perspective of what it’s going to take for those companies to develop algorithms.
Lumea doesn’t develop diagnostic algorithms. We chose not to do that and, and running a medical services company that looks out for the patients, that gives me another perspective and, and I’m realizing those are, there’s those, well, we all want to see. The data used for the benefit of the patients, and we all recognize that it’s probably going to take companies or large institutions to be able to support the development of those tools.
I don’t think we’ve really figured out how to do that data exchange in a way that is fair and ethical and that feels right. What I believe is needed is a is a trusted third party, and that’s how I came to the DDX Foundation. I can see the potential for these AI algorithms to transform the quality of care that we can provide, and I’ve had enough experience.
Seeing some of the power of these algorithms when [00:45:00] applied in the right way to the right population. It’s amazing. And so I fully believe that AI will transform the way we care for patients, how we get from here to there. We still have a lot to figure out. One of the problems is handling the exchange of data well.
And I think it’s impossible for the company that’s going to benefit from that to do it by themselves. I think it’s impossible for the one, for the aggregators of that data, the healthcare service providers to do it by themselves is there needs to be a trusted third party. That sort of manages this and that’s, and that, and that’s what I, you know, I’m trying to do with the DDX foundation.
So while I’m putting my, I’m going to, I’m putting, I’m selling, I’m kind of doing what I did 10 years ago when I bought the scanner, I’m going to sell a significant portion of my stocks and I’m going to donate [00:46:00] those to the foundation. To fund it for the first two years while we find those who can be trusted to.
to, to manage and oversee this, this foundation. So let’s take advantage of this digital, uh, YouTube video. Like you took advantage of the Lumia one. Who would you want those people to be? Like if somebody was to call your phone after this podcast, To support you in the foundation work, who do you want, what characteristics, what qualities do those people parties should have to be a partner for you to work with and like, and what’s in general, like the, let’s say value proposition for the world other than, yeah, and for them other than just, you know, being honest, um, Yeah.
And being fair, which is a good one in its own, but, um, it usually requires money. So let’s talk about that. Well, and so where does the money come from? Uh, maybe I’ll, maybe I’ll start with [00:47:00] that and then I’ll come back to, I’m going to answer this in reverse. How do you sustain this? I think there is, because there is a desire and a belief that these things are going to be valuable and who’s going to benefit from this, it’s going to be, besides the patients, it’s going to be the life science companies, the pharmaceutical companies, they’re the ones that are going to, I think, benefit ultimately financially from this.
And so, ultimately, I expect that much of the funding will come from. Those who are going to benefit financially. That said, those who can benefit financially cannot or should not be able to be the ones governing this if it’s going to remain a trusted entity. So if I put myself as conflicted, which I am, I’m conflicted as a healthcare service provider in that I’m conflicted as a technology company owner.
Ultimately I can’t be in that, that governing body. If it’s going to be trusted, not that you can’t trust me. But I have conflicts of interest inherently. [00:48:00] So, so ultimately what we see is a body that is going to be governed by a group that is universally trusted and supported one level down financially by those who are going to benefit from this and those who, and so the rule for this interplay, this exchange between life science, pharma, big technology, and healthcare service providers.
The rules for that engagement are going to happen in the context of what this group is doing, is outlining. But, you know, I think it’s a little bit like the open source community. Um, there are open source communities that are nonprofits that define sort of the rules for building this open source software that, you know, why, why does it exist because they realize that there’s value there that is good for humanity and should not be.
Necessarily controlled by any one company. And that’s sort of the realm where I see this foundation. Why would pharma choose to support it? Because I think if you can, [00:49:00] if not to get sued, well, not to get sued, but I think if we can build a system that enables them to access that data more efficiently, less expensively.
than they currently can because currently it’s very hard to access that because right now healthcare institutions, each one is hoarding it thinking, Hey, we’re going to make so much money out of this data. Honestly, the, the, the, the, the, the, I’m getting a little controversial now, but most of the big health systems, they all have big data plays and they all are sort of secretive about their big data play.
And then they all point to the other guys and they all say, Oh, those guys are selling their data while they’re doing, you know, everyone’s all the big systems are doing that the thing. The reason I think this has a chance at being different is because the groups that we’re working with are so small and so insignificant in the past.
Nobody has cared about getting their data, but in aggregate it exceeds any health system to patients that are in these. Yeah. [00:50:00] communities that we serve are the patients that are in ethnically underserved ethnic groups. Which is a big bottleneck right now in development of all those models. Like you mentioned, the big centers do not have access to those minority groups.
Exactly. So my belief is that if we can do this right, providing data that is less expensive than what they get from the centers, more a cross section of our patient populations that it will bring people to the table. And so we’re going to start off, um, just sort of trying to lead by example, doing some pilot projects and aligning the pharmaceutical companies and the life science companies who want access to this data.
The clinics will benefit for the work that they’re doing. Not, not, they’re not buying data. They’re going to do work to be able to provide the outcomes. For this will ultimately be less expensive, more efficient, more [00:51:00] standardized than anything that I’m seeing today in most data aggregation efforts. And the hope is that as we prove this concept, I’ll step away from being, you know, sort of leading it.
And so you just have like two years, right? That’s your time. Two years, two years to do this. Yeah. And this is where I’m putting my focus right now. And it’s a risk. But it’s, it’s really, it’s needed. And ultimately I know that there are smarter bioethicists, smarter technical folks, you know, there’s going to be people in that governing board that are going to be far more sophisticated than I am.
In, in thinking about the way this should be managed and my goal is to find those people and then create a board that is trusted and then come down and try to work in the, you know, in, in, in, in this exchange process so that we can actually start providing, start bridging this gap between where healthcare and [00:52:00] technology need to converge.
So that’s, that’s a lot of heady talk, but I, hopefully it makes sense. It does. And let me tell you, so it’s not the first time, not exactly in the same way, but it’s not the first time I hear this concept. There is a public private consortium going on in Europe right now called the big picture. And they’re Right.
You know about that one and they haven’t one of their packages. One of the work packages that they have is called honest broker because their goal is also to aggregate a lot of image data to give access to this data to for entities to develop algorithm on. But they need an honest broker for people to provide and for people to extract the data from.
So it kind of is something. Yeah, ultimately, I know we’re going to have to bring whatever we’re doing together with like minded like that, and I love big, I love what big picture is [00:53:00] doing. I am a fan of Big Picture. I’m going to have them on the podcast again in September, uh, because they’re meeting for, for discussing, uh, in Dublin in September.
So I’m going to be there. So fantastic. This is amazing. That’s awesome. A cool story. I know. Fantastic. Thank you so much for joining me. And okay, so I’m going to be on the lookout for the outcomes in the next two years. Judging by your other endeavors, two years is the timeline to build up and exit, right?
I’m curious what’s going to be your next thing. I just want to go back and practice pathology. Yeah, I do. Oh my goodness. I might not take me again. Maybe, maybe. Um, so another question, just one last question before we finish. What kind of data do you guys want to aggregate in the foundation? Are we talking images as well?
Everything? [00:54:00] Yeah, so we have we have over, so if you just look at what we have access to today through path net and other clients, we have over a million prostate biopsy images. Uh, probably, I mean, it’s probably one of the largest and they’re all standardized. They’re all done in three labs, similar protocols, same scanners.
So it will be a very, I think, you know, it’s all, it’s already. The, the challenge is not actually, we’ve overcome the challenge of aggregating a massive amount of image metadata. The foundation will focus on consent and, uh, putting those kinds of
things in process that will, uh, be sort of an independent, it will act sort of like an IRB. And some of the things that large institutions have in place, which are out of reach for community practices. There’s a small. So essentially our, our goal is to aggregate most, it’ll mostly in the first year or two, it’s going to be focused on prostate cancer because [00:55:00] that’s.
Because that’s our, that’s what, that’s where we have the data. But as you know, as you know, if this is successful, then we’ll venture into other things. Okay. Thank you so much for joining me, Matt, and I wish you a fantastic day. Thank you. Great to, great to be with you. Thank you so much for tuning in. If you stay till the end, it means you are a true digital pathology trailblazer.
And I have something special for you. Recently, we have launched the digital pathology club membership. It’s a library of all our courses. covering image analysis, AI and pathology, digital pathology, lead generation, and we are creating new courses every month. After each course, you get a certificate of completion and you can easily keep track of your continuous education.
And I would like to give you a free trial of the membership so that you can see what’s inside yourself and be in the know about the digital pathology principles. And the new [00:56:00] advancement. So go ahead, click the link in the show notes and join me for a free trial of that digital pathology club. And I talk to you in the next episode.
