The Group Dentistry Now Show: The Voice of the DSO Industry

Our podcast series brings you dental support and emerging dental group practice analysis, conversation, trends, news and events. Listen to leaders in the DSO and emerging dental group space talk about their challenges, successes, and the future of group dentistry.

In this episode, Ophir Tanz and Dr. Kyle Stanley discuss how Pearl is revolutionizing the dental industry. Dental groups and DSOs with the best equipment, data and analysis have the edge, but many dentists are still analyzing x-rays manually. Pearl’s AI technology identifies dental pathologies diagnosed in x-rays with three times the accuracy of the average dentist. Listen to this podcast to learn more about the future of technology enhanced dental care, including Pearl’s new “Second Opinion” technology. The two guests also give their opinions about the future of AI in the dental industry.

The Group Dentistry Now show has listeners across the North America, Australia, Europe, South America and Asia. If you like our show, tell a friend or a colleague.

Watch the video interview or choose your favorite listening app below and subscribe today so you don’t miss an episode! Full transcript is also provided below.

Choose your favorite listening app and click on it for a direct link to the show:

FULL TRANSCRIPT

Bill Neumann:

I like to welcome everybody to the Group Dentistry Now show appreciate y’all coming back for the second episode of 2020 and I am your host Bill Neumann, and we’ve got two gentlemen from probably one of the most disruptive companies in the dental industry for the new decade. AI is all the rage. I don’t think a lot of people even understand what AI is. So maybe the gentlemen from Pearl can explain what that is, how it’s going to impact DSOs, the ones that you work in, and then also how it impacts the industry as a whole. So first off, I’m going to introduce Dr Kyle Stanley and then Ophir Tanz. So these two guys are with Pearl, and they are really doing some creative things. There was just an article in the Wall Street Journal today, so they’re going to talk a little bit about that and how they are helping insurance companies.

Bill Neumann:

So Dr Kyle Stanley, he’s named as the the next generation of cosmetic dentists by the AACD and a top 10 young educator in dentistry by the Seattle Study Club. Graduated from the Herman Ostrow School of Dentistry of USC, completed the dental implant residency along with the dental implant specialty in Brazil. He’s also a dedicated researcher who has published and has been published in some of the top international dental journals, topics related to aesthetics, dental implants, and smile design. He’s also a key opinion leader for many well-respected companies and has presented on five continents and is a high in demand speaker. I think you spoke in Canada, an event that I was at just recently in November, but missed you.

Bill Neumann:

And I believe one of you two is also going to be speaking at an event right down the road from where you’re located in Beverly Hills in April. I think it’s the DSO Leadership Summit, correct?

Ophir Tanz:

Oh yeah, I think that’s me.

Bill Neumann:

Yeah. I think that’s you. Yeah, I just found that out yesterday. You’re all over the place. Both of you. With Dr Stanley’s company Pearl, he is changing the way patients are treated through artificial intelligence and as a leader in the field. And Dr. Stanley actually has a practice in Beverly Hills. All right, so that’s enough on Dr Stanley right now. Ophir Tanz, serial entrepreneur, looking at this bio here. Prior to founding Pearl to introduce AI to the dental category, Ophir was the founder and CEO of GumGum, an applied computer vision company focused on the media category that he grew to hundreds of millions of revenue. Prior to GumGum, he was the CEO and founder of MoJungle, a mobile media sharing platform that was sold in 2007. And before that he co-founded Fluid Design, an interactive and branding agency.

Bill Neumann:

He has a BS and MS from Carnegie Mellon, right down the road from us here, and he’s an active member of the LA startup community serving as an investor, mentor, and advisor. You guys are overachievers. Well welcome to the Group Dentistry Now show.

Dr. Kyle Stanley:

Thanks for having us.

Ophir Tanz:

Yeah, it’s great to be here.

Bill Neumann:

Yeah. Good stuff. So why don’t we talk a little bit about Pearl, if you can give the audience an introduction to the company, who Pearl is, what you’re doing, and then maybe a little bit of an explanation as to AI, because I know it can mean so many things. Maybe what AI means to Pearl and then what it means to the dental industry.

Ophir Tanz:

Yeah, absolutely. I’m happy to take that. A little bit of context behind Pearl. I started GumGum, which you just mentioned in the bio, back in 2008. That company was focused primarily on a technology called computer vision, which is a subset of artificial intelligence, and that particular capability is focused on teaching computers to see and understand the world, and obviously are able to perform a variety of tasks and analyses based on that understanding. So at GumGum, as you mentioned, we apply that technology widely, primarily in the media category, but it was at GumGum that we had the idea to apply this technology to healthcare and specifically to dentistry. So as far back as three and a half years ago now, we actually started the collection of the data assets that would ultimately be required to train neural networks, which primarily drive our capabilities in computer vision. And we’ve collected, to our knowledge the largest collection of dental radiographs and also intraoral scans in the world.

Ophir Tanz:

Kyle and I started down that path several years ago. I was a patient at his office, and that’s how we had come to meet. Ultimately we became excited enough about the opportunity and the impact that we’d be able to have on the category and felt that this technology would be able to more effectively achieve its full potential as a standalone entity. So in May of last year we spun GumGum Dental, which is now known as Pearl, out into its own company and we’ve been operating as an independent entity. I also stepped down as my role over an 11 year period operating GumGum as the CEO in order to come over and run Pearl, because I’m incredibly excited about what we’re going to be able to accomplish in this category.

Bill Neumann:

Well, that sounds exciting.

Ophir Tanz:

And I should say, one more thing, just to put a finer point on it all, the focus of Pearl is to apply artificial intelligence and specifically computer vision to imaging challenges in the category of dentistry. We’re really touching every major constituent in this category. I’m sure we’re going to talk about that a little bit later on, but that’s the primary goal, the mission of the company.

Bill Neumann:

Okay, great. I appreciate that. That helps, I think, the audience out a lot. It certainly helps me out. So why don’t we talk a little bit about AI and Pearl of course, and how do you think that’ll specifically impact the DSO space? The DSO space meaning, could be the large platforms with hundreds of locations, but also the impact it would have on emerging groups. Multisite, but maybe smaller doctor owned, doctor led groups.

Ophir Tanz:

Yeah, I’m happy to start with that as well. If you look at the sort of data that exists in the dental space, whether that’s an intraoral scan or an insurance claim or a set of radiographs or even… A lot of PMS, a lot of that data is unstructured, meaning that it’s not a searchable, and if it’s not searchable, it’s not really usable at scale. One of the key things that we do is that we’re able to look at any set of radiographs, for example, whether it be a thousand or a million and develop a deep pathological understanding of the contents of those radiographs and of that data, and actually break out, for example, patient populations by pathology and a number of other factors. So really overnight you can go from having a very cursory or anecdotal understanding of a large group of patients to a very specific one which enables you to do things like codify treatment plans and communicate with patients and deliver quality services at scale.

Ophir Tanz:

So one of the ways in which we feel we’re going to impact the DSO category is with a product that we have called Practice Intelligence. And Practice Intelligence is a set of tools that enable DSO managers and practitioners to interact with patient data in a way that is very clarifying and very powerful. There’s also other capabilities like being able to review insurance claims that have been submitted, to review intraoral scans, to understand the quality of preparations at the level of the prep tooth, and ultimately to deploy critical training in very surgical ways to elements within the DSO so that the overall quality of care could be improved very quickly.

Bill Neumann:

Excellent. So that is Practice Intelligence.

Ophir Tanz:

That is practice intelligence primarily, yes.

Bill Neumann:

You can look at it from a training standpoint. So you can take a look at the skill set, maybe dentists have in a particular DSO and even benchmark dentist against dentist, I guess, and see-

Ophir Tanz:

Yeah, so you can rank order dentist against dentist and you can do so in highly objective ways, again, that are very clarifying and enable you to deploy the appropriate training and protocols to very quickly improve the quality of patient care.

Dr. Kyle Stanley:

And whether that is through radiographs or even like Ophir said, through intraoral scans, we can rank the dentists. How good are they prepping? Are the margins on their preps able to have crowns made consistently on them? Are they doing the correct treatment based off of what is found in the radiographs, so that the DSO managers can then make the decisions? Who needs training? Who needs a raise? Who needs to be fired? Really understand the data-driven aspect of managing them.

Ophir Tanz:

Yeah, I mean intraoral scans are a good example just because they’re so straightforward. If you look at the vast majority of intraoral scans that are being captured for the purposes of restorations in market, the vast majority, over 80% we found are suboptimal. If you have a sub optimal prep, you’re going to have a suboptimal restoration. There’s just no other way around it. This is something that can be taught, but is routinely being done incorrectly or just again not optimally. We’re able to surface that immediately. This is something that doesn’t require anybody to be fired, isn’t on that big of a deal, you can literally sit a practitioner down and show them how to do it better, and they immediately will do it better. And then you can actually validate that they’re doing better in a move forward fashion. So automatically the quality of the restorations is going to increase across your DSO because you’re providing the lab with higher quality models.

Bill Neumann:

Right. Makes a lot of sense. You didn’t mention this, but tell me a little bit about Second Opinion. So tell me what that is, how it impacts the clinicians and then also the patients.

Ophir Tanz:

So Second Opinion is the technology that we built that is able to do real time pathological and non pathological detection on radiographs. So bite wings, periapical, panos, on and on. It works in a very granular fashion. In other words it identifies types of caries, types of margin discrepancies, periodontal ligaments, bone loss, calculus, the doctor knows better than I do. But the list goes on and on and we have also proven with independent studies that these capabilities are operating in the vast, vast majority of cases that are either equals to or better than human levels, and this technology we anticipate later on this year will be integrated into the existing workflows of dental practices so that when a radiograph is captured, we’ll be able to immediately, in real time, overlay the detections that are made.

Ophir Tanz:

It’s called Second Opinion intentionally because we’re providing a second set of eyes to practitioners. They’re still in charge. The call is still theirs to make, but with small forms of calculus or bone loss or caries, these things are easy to miss. The detection rates tend to vary dramatically and things like mood and whether or not you have to go to lunch in 20 minutes, all these factors could affect a dentist’s ability to read radiographs. And that’s really a critical component, obviously, of providing a high quality care.

Bill Neumann:

So Second Opinion is dentist takes a radiograph and then through the database of millions of… The history that you have already built in that you can take a look and based on the radiographs that’s sitting in front of the dentist, and that library of information, you can overlay that information over the existing radiograph and there’s some predictability behind that. You can say, “Based on all this history, this is what we see should be done?”

Ophir Tanz:

Yeah. That’s really where the AI comes in. So we’ve trained these systems to identify pathologies, and when encountering a new radiograph, it’s able to use that wealth of knowledge that’s been trained for years over tens of millions of radiographs to identify pathologies, that has never seen that particular x-ray before, but that it’s seen many times over.

Bill Neumann:

So next question. What about the FDA? Does there need to be FDA approval? It seems like there may with something like this.

Ophir Tanz:

Yeah, so we have a number of products. We have insurance products, we have laboratory products, we have DSO products, and then we have Second Opinion as well, which we just discussed. Second Opinion is the product that does require FDA approval. We’re currently pretty far along in the clinical trials phase and expect to have that completed and hopefully granted in short order. And the reason that component requires FDA approvals because that pathological detection is patient facing in real time, and if practitioners are making treatment decisions driven by the technology, it’s very important that that technology be qualified to be accurate so that the appropriate treatments are being deployed. For a lot of the other technologies that we have working behind the scenes solving and automating a variety of problems, the FDA approval is not particularly relevant to that sets of products.

Bill Neumann:

Okay. Let’s pivot away from SSOs for a second. First off, congratulations on the Wall Street Journal article. The timing’s perfect. I saw that this morning and it really leads into this next question. Besides DSOs, what other elements of the dental industry is Pearl relevant for? Obviously in the Wall Street Journal article you talk a little bit about your work with insurance companies, but that’s one element besides DSO, so why don’t we talk about that?

Dr. Kyle Stanley:

Yeah, I can talk about that. So we have our clinical aspects, let’s say, with our Second Opinion and then we have the business aspect with Practice Intelligence. The other two areas of the dental space that we’re dealing with are insurance companies like you said, and the other one is labs. I’ll start with labs. When we analyze intraoral scans, we can really automate things for the labs. So many processes in the labs are very step by step, and even solving one step can make it much more consistent, much more reproducible, much less costly for the labs. So when we are analyzing our intraoral scans and giving the scan a score, we’re now giving our labs a way to know which scan should go to which particular area of their lab, but also how their doctors are doing, so they can get an idea of who are their best doctors, who are their worst sectors, let’s say.

Dr. Kyle Stanley:

Also, once we can take that scan and analyze it, we can mark a margin on it, and this is something that is happening over and over and over again, and some of the largest labs in the world that we work with have many, many people that all they do all day long is mark margins. When a human is behaving like a robot, it’s better to have them behave like a human. So we can take those people from marking margins, have the AI do it consistently, and move them to something that really requires the human aspect, the artistic aspect. The staining, glazing, final polishing, customer service. So it becomes a way of our labs to automate some of their business, for the doctors to know that they’re going to get something very consistent every time, it’s the same engine that’s going to be marking the margins. That’s the lab side.

Ophir Tanz:

I was just going to say as a general note, we really do believe that this capability is going to affect every constituent in the category. And this isn’t something that’s really unique to Pearl and dentistry. This is something that is unique to artificial intelligence in medicine. If you have the capability to improve patient care, to ingest massive quantities of data, whether that be semantic textual data, or imaging data or patterns over periods of time for big data analysis, then you will have the ability to do things better. The stakes are very high in this category. The costs are very high in this category. The economics matter, the care matters, and what we see is just a huge amount of opportunity to streamline so many aspects of this space. So as it relates to practitioners pretty straight forward what the value proposition is there and how that grows over time with Second Opinion and treatment planning and doing that for more forms of imaging and other integrations of AI.

Ophir Tanz:

On the insurance side it’s not only about the fraud, waste and abuse detection, which the Wall Street Journal article talks about. Which by the way, we really do want to stress that we do think about the world from a dentist’s perspective. Kyle Stanley is obviously a dentist. My father is a retired dentist.

Dr. Kyle Stanley:

My dad and my brother are dentists.

Ophir Tanz:

And we feel that it’s unfortunate for a small set of bad actors to spoil and tarnish the reputation of everyone in a very, very large industry, the vast majority of whom are doing quality work and trying their best and not doing anything that could be considered nefarious or illegal. So the ability to deploy these technologies to fundamentally solve those problems, eliminate those issues on a move forward basis, also helps all the ships in the category rise.

Bill Neumann:

Yeah, absolutely. It’s not just about preventing fraud, waste and abuse. The assumption isn’t that that’s the majority, that’s the minority and to have this safety in place almost where you prevent… People won’t be doing this or even thinking of doing this because they know now that they’re just being monitored.

Ophir Tanz:

I’ll mention two other quick things, just because I think they’re relevant to this point. They’re a little bit longer term initiatives, but one of them is we’re working with a number of universities to develop educational tools. Those educational tools are going to help train dental students to more accurately read dental radiographs, which coming out of school, almost everybody is subpar. We have the head of all these radiology departments, the dentist schools, they all freely admit that. They wish there was a better way to more consistently train dentists to read radiographs so they don’t have to learn on the job over a period of five years or 10 years or 20 years.

Ophir Tanz:

Other things that we’re doing around the intraoral scan side and prepping. The ability to leverage this technology, to deploy real time training and feedback, we think will produce better dentists out the gates. Another thing that we’re doing is working on creating studies and connections between oral health and systemic health, which we believe over the next 10, 20 year period is going to be… There are major discoveries there that are ready to happen, but we just need a lot more research and data to establish those correlations. There’s already interesting research to that end, but that’s also a space in which we intend to continue to play.

Bill Neumann:

Excellent. Yeah, it makes a lot of sense with companies, DSO’s like Pacific Dental becoming more ingrained with using Epic now and opening up healthcare… I think they have one in primary care in Nevada, you have Aspen Dentals has some urgent care clinics in Syracuse. You have Walmart Health. That certainly seems like a great space to go into, to get more tied together. And then to your point also on the mentorship and the educating, some of maybe the younger dentists read radiographs, we have a lot of DSOs that have younger dentists, so that makes a lot of sense there. And one of the struggles they have is how do we educate? And we’re not educating one clinician, we’re educating a thousand clinicians. So how do we do that in a scalable fashion? So let’s go over this again real quickly. Products that fit into the DSO space. We mentioned Second Opinion a little bit, but we also had Practice Intelligence as well, right?

Ophir Tanz:

They really all fit into this space. So you have a Second Opinion for realtime analysis, Practice Intelligence for much more in depth and powerful analysis of patient populations and things like treatment planning and codifying that. You have the intraoral scanning suite, which we call Smart Margin, which if they do in house milling, we’re able to draw margins autonomously and also the ability to score preps both historically and in real time for the purposes that we had mentioned, and also the ability to do both retroactive real time fraud detection on the DSO side. So that’s a product that is not solely intended to be used by insurance companies. It’s also intended to be used by industry so that it can identify patterns of fraud, waste, and abuse within their own entities and get ahead of those problems.

Dr. Kyle Stanley:

Before it happens.

Bill Neumann:

Yeah, that makes a lot of sense. So let’s talk a little bit about… I think I have time for one more question. You talked a little bit about it already, Ophir. What’s next for AI in dentistry? Maybe it’s not even pro related, but AI in general. What do you see? What are you doing? You talked about the oral systemic connection and focusing more on that. What else?

Dr. Kyle Stanley:

I can expand on that a little bit. I think that a lot of the mundane daily routine things will be automated with AI. I think that we’re going to be very close to where there will not be a front office staff, that will all be driven with natural language processing. N one’s going to be answering the phone. It will, again, all be done with scheduling through an AI application. Getting into more for the surgical aspect. I think all implants will be planned, milled, printed through AI, meaning that we’ll take STL files, the intraoral scans, bring everything together. The AI will decide where the best place is to put the implant, how deep it should be, which implant it should be, automatically generate an STL file and have that being milled. So I think anything that is mundane and tasked to the dentists more on the back end will be automated.

Dr. Kyle Stanley:

Talking to the computer to perio chart instead of writing it down. I don’t think that we are close at all to the clinical aspect being completed by AI. I think a lot of dentists are worried when they talk to me like, “Are you guys going to take my job?” We’re not there. But what we’re going to do is make the profession more consistent, more reproducible, more… I think dentists will make more money because we can reduce overhead oftentimes, and I think it will be a more enjoyable profession. Right now dentistry can be a stressful profession and when you take some of these mundane aspects, when you reduce the risk of a dentist or a dentist owner, it becomes a very enjoyable profession.

Ophir Tanz:

I would just say from a purely a technological perspective, there is a field obviously called artificial intelligence. That field unto itself is advancing rapidly. There are a lot of breakthroughs that need to take place until we have anything that resembles what people oftentimes think of when they think of AI or robots. We’re very far from that period, but advances are being made at a rapid pace. For example, today we talked about these very large data sets that need to be collected in order to train to identify pathologies or mark margins or do any of these tasks that will be talked about, and then there’s a lot of additional pre and post processing that needs to occur. One example of a technology that is getting more and more powerful and capable, still little bit of a ways away, is something called unsupervised learning where you can deploy an intelligent agent across a set of much smaller data and it can come out with, in the future, presumably, a lot of the same types of capabilities, and it doesn’t need to have its hand held by humans in the way that it currently does.

Ophir Tanz:

So it’s really just the first inning as far as this technology is concerned, and as the technology continues to advance, we’ll be able to continue to apply that to this category, hopefully in a very exciting and effective ways.

Bill Neumann:

Lot more to look forward to in the AI category and it’s not necessarily something to be afraid of but should be embraced in the idea is it’s going to make dentists more productive, more focused on what they can do clinically, better clinicians. So that’s great. Well, listen, I think that’s about all the time we have today, but I really would like to thank you guys for talking about AI in general, about company Pearl. It’s exciting stuff. I think it’s going to be super disruptive in a good way and really helpful for clinicians and then also the DSO industry. It sounds like you both are going to be on that speaker circuit at several DSO events, so we’ll definitely see you around and get to hear what new and incredible you have in store for us. Thanks Ophir, thanks Dr Stanley. Really appreciate it today.

Ophir Tanz:

Okay.

Dr. Kyle Stanley:

Take care. Bye.