Really, really nice to to have everybody here.

Today, I'm going to be talking about artificial intelligence, what we do at Pearl, together with our collaboration with SOTA. So it's a really, really exciting time.

And I'll be answering questions at the end, so feel free to put them in the Q and A. I'll go through those at the end.

But let's get this party started.

So, I'm going go over some basics of AI.

We're going to go into how AI is being used in radiology, and then also specifically what we're doing with SOTA Cloud.

So, let's go. Here we go.

My name is Doctor. Kyle Stanley, and I am the Chief Clinical Officer here at Pearl. I'm also the co founder of Pearl. I am a practicing dentist.

And today we're going to be looking at what are you missing out on if you're not using AI, right? Because we have a lot of technology in dentistry and you hear about AI all the time. But what are we missing out on, right?

So a little bit about my history is that I have a fairly interesting view of what's happening in dentistry because I've taught at university for nine years, I have a private practice, I've done research.

You know, we started Pearl many years ago. I worked in a laboratory. So I have this kind of global view of dentistry.

And I think the question for today is how can AI help you in your practice today?

Not like in five years or in ten years, everyone kind of knows this is the future, but how can it help you in your practice today?

And I think when you look at when you look at technology in dentistry, you can feel a little overwhelmed. Everything is coming at you. It's like I have to learn aligners and digital bite sensors and guided preparation and chairside milling and digital jaw tracking and lithium disilicate and MarPI and SFOT. And it's just, you know, so much new innovation.

And then there's an update, and then there's a new milling machine that comes out, or there's a new scanner. And AI is different in that case. And the reason why is because AI is plug and play.

There are immediate benefits.

It is simple. It's time saving and a relatively low investment.

And so when you look at dentistry, we have some problems in dentistry. Let's be honest, my dad's a dentist, my brother's a dentist, but we have inconsistency and inefficiencies.

We have a ton of untapped data that we're not utilizing.

And we, of course, have unexplored business opportunities, opportunities where we can grow our business by utilizing technology. And so today, in our short time, I hope to show you a little bit about how we can take these problems in dentistry and really make them with the help of AI into consistent and efficient decisions, benefiting from your data, maximizing business opportunities, and do that with low investment and easy implementation.

So a quick little definition of what AI is, because we hear it all the time, but what does it really mean?

AI is computer systems that can simulate the problem solving and decision making capability of the human mind. So a machine that is making decisions just like us humans. And I think most of us, when we think of AI, we think of like this. We think of general intelligence, which is like, the Terminator.

You know, this all encompassing, training itself and reprogramming itself faster than humans can do and taking over the world and whatnot.

And that is not what we have today.

No. Most AI experts agree that we do not have general intelligence, and we may never have general intelligence, actually. What we have today is called narrow intelligence.

Narrow meaning it does one thing and it does that one thing very well.

So for example, you know, recognizing an apple in an image. It does that one thing, and it does that one thing very well. It's not gonna do anything else except for recognize that apple. Right?

And when you look at the history of of modern AI, it's been around for a long time. I mean, since the nineteen fifties is when AI started. However, it wasn't until personal computing and the dot com boom and the rise of the Internet that we had the building blocks that we really need for true modern AI. So what we needed was data driven algorithms, which, you know, we've had for many years.

But what we didn't have was the processing power. So, you know, in your, you know, current phone, it's much faster than the phone that you had five years ago and a few years before that.

There's something called Moore's Law, which says that the processing power doubles every eighteen months and halves in cost. I don't know why our phones keep getting more expensive, but, the processing power is definitely getting faster.

And also big data. Even if we had data driven algorithms, processing power, and we didn't have big data, think about your practice twenty years ago if you've been practicing that long. We didn't have digital charting. We didn't have digital radiographs.

We didn't have, digital impressions. We didn't have CBCTs. Now we have all of this data, but we're not doing anything with it. And that is where AI is gonna come and bear fruit for all of us.

And you've probably seen different types of applications of AI, in your everyday life, whether it's predictive analytics, which is like, you know, bank alerts or even navigation. I mean, I use navigation even when I know I'm going somewhere.

Or natural language processing. You've definitely seen this with ChatGPT, with Siri and Alexa and chatbots. What we're gonna talk about today is computer vision.

This you've seen in in Teslas and you've seen in manufacturing and, you know, when you come back through global entry and it recognizes your face and you go right through, that's allowing the computer to see and interpret imagery just like how a human would.

So when you look at the history of modern dental innovation, you know, the the profession has moved forward because of these innovations. In the last twenty years, have things like digital dentistry, intraoral scanners, aligners. And I really think that that next big innovation is artificial intelligence.

And that soon, AI will be an essential utility in dentistry. And I say utility just like how you think of water or power, something that you can't imagine practicing without. I think in a few years, we'll look back and say, can you believe we didn't have AI in all of our radiographs or AI in our treatment planning or AI in, you know, our intraoral scanning?

And so what I want to go through today is really AI for consistent diagnosis.

So we have gone from the night to the day.

So you think of of modern radiology, and we're still looking at radiographs in this, black and white or grayscale. And that may be okay for us.

But what about showing it to our patients and letting them understand what's happening?

I think it's much easier to communicate with our patients when they can see what we can see, for example.

So let me show you a reason why we need AI. Because you may be thinking, why do we need this? You know, my dad practiced for forty five years, and he didn't have AI. And he said, you know, why do we need this, Kyle? And I said, dad, could you have done it better?

And he said, Yeah, you're right. I could have.

So let's look at these two studies. On the left side, you see a study from the Dental AI Council. And what was done in this study was that they gave one hundred and thirty six doctors an FMX, full mouth series of radiographs, and said diagnose and treatment plan this patient just based on the radiographs.

And what you see is a huge difference in what was treatment plan. So for example, on an upper molar, you have a fiftyfifty split on whether there was caries or not.

And then when you go to the entire mouth treatment plan, it ranged from three hundred dollars to thirty six thousand dollars.

So this is like you come to me and I say, yeah, you need a filling. You know, it's three hundred bucks. You go to my brother and he says, you know, we need to do some crowns and, you know, some root canals, scaling a root plane. It's gonna be about ten thousand. Then you go to my dad and he says, oh, it's thirty six thousand. We need to do a full mouth rehabilitation. This is really happening in our profession.

Now, a study that we did on the right side with Pearl is that, we took three doctors. So one was a maxillofacial radiologist and the other two were general dentists. And we just asked them, Where do you see caries in these, you know, almost nine thousand images? They only agreed four percent of the time.

Now, is between other doctors, but also with ourselves.

We only agree with ourselves about seventy percent of the time.

So if you or I look at a radiograph, let's say we look at an FMX and we diagnose it, we say what we see, and then we look at it in three months or in six months or a year or whatever, we're only going to agree with ourselves seventy percent of the time. Now why is that? It's because us humans are inherently inconsistent. If you think about it, some days you have your coffee, some days you don't.

Sometimes, you know, your kids slept, sometimes you don't. I have a nine month old daughter, so most of time she doesn't sleep. Sometimes there's traffic, sometimes there aren't. Sometimes you get lunch, sometimes you don't.

So we are, you know, never really consistent as humans.

And so, adding in something that is inherently consistent will definitely help us.

So, we can go from inconsistency now talking about this untapped data that we have. You know, when you think about how we've painstakingly digitized our entire practices practices. That's taken a lot of time and money, but we're not benefiting from that data. It's just sitting there.

And now it's time for AI to come in and really help with that.

And the last thing is unexplored business opportunities. You know, most of us went to dental school to be dentists, not to be business owners, and we get out of school and we're business owners. So we do lack profits. We have poor patient experience and we have limited growth. And so, this is definitely something that we can improve on.

I love this quote.

I not only use all of the brains that I have, but I can borrow.

And that's exactly how we need to think about AI in radiographs. So us dentists, we're pretty good at reading radiographs. Right? The AI is pretty good at reading radiographs. When you add the two of us together, we become better than the human.

So the human plus the AI is always better than the human by themselves or the AI by themselves sometimes.

So AI is consistent, inherently inconsistent, just like we are inherently inconsistent. It is accurate. By no means is it perfect, but it is accurate. It is objective and it is fast.

So now when you take a radiograph, what you will see within your SOTA Cloud imaging, and I'm going to show you this later when I go directly into the imaging software, is that you won't see black and white images anymore. You'll see images that have caries and bone loss and calculus and periapical radiolucencies and margin discrepancies so that you can really, when you walk into an operatory and you see an image, it's very simple for your eyes to go to the area of interest.

So, it decreases your decision fatigue as well.

Instead of kind of searching around everything, the AI will point you in areas and then you can go in and investigate further.

And again, we're going go in and I'm going do a demo of this, both in Pearl and also in soda to really show you the difference.

And this is some data from our FDA trials. What you see is that when a dentist is utilizing artificial intelligence, they're able to detect thirty seven percent more disease.

This is disease that was previously going undetected. Now it's not always something huge. It may be a small amount of, bone changes from the CEJ to the bone measurement. It may be, you know, small interproximal decay. You may not even treat it, but you can detect it and talk to the patient about their diet and their flossing. And maybe you want to remineralize it using peptides or using, some other type of treatment to remineralize it.

But you can detect it, and therefore, you can monitor it and maybe treat it. But I love the graph on the right because the graph on the right really shows how we need to think about artificial intelligence.

Like I said, us dentists in the orange, we're pretty good at reading radiographs. Right? We're pretty good. By no means are we perfect. The AI is pretty good at reading radiographs. By no means is it perfect. But when you add the two of us together, we become superhuman.

We become better than ourselves.

And what's great about the AI is that it's there as your assistive tool. It's not there to take over anything that you're doing. It's there to, you know, increase your efficiency, to build trust with your patients. We'll talk about that a little bit later.

To decrease your decision fatigue, to increase your sensitivity.

So there may be things that the AI shows you and you say, I don't really know. Is that is that not caries? And you can go in and investigate and say, oh, it is or it isn't. But again, it's there to point out areas of interest so that you can go in and make that final decision.

And it's constantly improving. I mean, AI, every few weeks, something comes out and it becomes better.

So let me go in and show you a few more things.

So one thing we also need to think about with artificial intelligence is building perceived value.

And this is very important with any type of business, but especially in dentistry, where our patients come in knowing that we have financially driven diagnosis. What does that mean? It means the more I diagnose, the more money I potentially make. Right? And that's just how it is. It's like a mechanic or it's like many other jobs.

But they know that if we diagnose more, we make more money. And that doesn't look good for us. That decreases the trust that happens with our patients. And when you look at the studies involved with this, it shows that, you know, fifty five percent of patients say trust in the diagnosis or in the dentist is the most important factor in treatment acceptance.

That is huge. Trust. Not the cost. It's trust.

Sixty four percent of patients say they don't understand their X rays during a consultation.

So this happens many times, and I'll show you in the demo where we go through in black and white, we're saying, Mrs. Jones, you see this tiny little thing here?

And, they're going, What? I don't know what you're talking about. Now you're going to see with the help of Pearl and SOTA together in your radiographs, you'll be able to show them with color and boxes and little polygons and lines and colors. They really understand what's happening.

Eighty seven percent of patients say it's important that their dentist uses the latest technology.

And this light blue circle, this is very important. Remember this one.

Seventy one percent of patients say they would trust their diagnosis more if their dentist is using AI.

Wow, that is very, very powerful. So, when we studied this, this was done with multiple offices from around the country. What we found was that the impact on treatment plan acceptance was massive. I mean, you look at crowns, how many crowns do you do a week if you're a general dentist? The treatment acceptance went up thirty four percent simply by showing the patient what's happening on the radiograph.

So if there's caries, if there's open margin, you know, it wouldn't change for aesthetics, of course, but showing pathologies on the radiographs will change your treatment acceptance treatment acceptance and really make it go up.

Scaling and root planning as well. You know, I know patients come into my practice and we before we had AI, we would say, you know, you need scaling root plan. And they say, well, I just want regular cleaning.

Like, well, you can't do that. You have calculus. You have bone loss, whatever. We may even show them the radiograph. They don't understand it.

Now, with the help of AI, we can go and show them the radiograph, show them the measurement that is unbiased. Right? The AI has no incentive to find more. It doesn't gain pleasure or get money from finding more disease. So it's really unbiased. And that's why we call it second opinion. It's a built in second opinion that is unbiased and there for the patient and the doctors.

So let me show you a few case studies on, this came from one of our clients in Florida.

And what they're doing is showing all of their patients what's happening with their AI detection. So they have a patient that comes in and said, I just want to get the tooth pulled. The doctor was able to use second opinion and show them the decay not only on the one that had the pain, but also on the adjacent tooth and show them, okay, you know, we shouldn't pull this tooth. We can save this tooth.

And what they were able to do was do a highly productive appointment, right, where you're doing endo and you're doing the adjacent resin and, you know, prepping the crown, but also save this patient's tooth. I mean, this patient was convinced they just wanted it out because it hurt. But they could show them how they still had so much tooth structure remaining even after doing the root canal and the crown. I mean, there's still going to be a much better prognosis of saving this tooth long term.

So I love this because from dentist perspective, you are doing great treatment, minimally invasive treatment, but also you're able to save this patient's tooth, and it's a productive day.

Let's look at this case. An elderly patient presented to the office simply wanting her free cleaning and no other work.

The office used second opinion to provide her with an AI driven visual walkthrough of all the conditions detected in her x rays. So this is the patient that comes in saying, don't want to do anything.

But you show them the calculus, the, bone loss on certain teeth. You showed them the caries, for example. And not only caries, but into enamel, into dentin, into pulp, into cementum, showing them how far it's progressed.

And I think you really gain trust with these patients. And look, it turns into a fairly productive appointment.

And the last case study that I'll show you is that this patient presented to the treatment or to the office as an emergency patient with a broken off lower molar, and they were able to show the broken piece with caries, with adjacent caries on the little interproximals of that second molar. And look, it turns into, you know, a fairly good restorative day and also a way to really treat this patient in a good way.

So what I love about dentist what I love about dentistry is is helping patients, but I hate looking like the bad guy. You know, they come in with pathology, and I see the pathology, and then I get in trouble. So what I like about this is I can say, you know, missus Jones, I I see I see caries here. I see decay, what I would say to them. But not only do I see this, this third party AI system that is unbiased agrees with me in this case.

And I want you to know that there's going to be times when you may not agree with the AI, and that's totally okay because you are the first opinion. That's why we call it second opinion.

So let me show you some things. First, I'm gonna show you traditionally how we had it in in Pearl. Right? So in Pearl, it's a separate window, and you can click on a patient.

And then you go in. I'm going to look at this image.

And so, here you can see I'm seeing calculus here. I'm seeing periapical radiolucency. I'm seeing some bone measurements that are above the American Academy of Periodontology, two point five millimeters. So it's showing in yellow.

And this is very, very nice for the patient to see, right? For your hygienist, think about showing them what's happening with calculus.

Or, this periapical radiolucency being automatically detected. When you walk into the room, you immediately see it or this bone level. Now where the magic happens is when you bring in tooth parts.

So tooth parts outlines every aspect of that image automatically, segments it and colorizes it.

So it becomes very easy to explain the depth of caries, for example, or, you know, where we measure bone from.

And you can really have this great experience with the patient.

Now let me show you in Because what's great about having it in SOTA is this is the best workflow to utilize AI.

You know, our in dentistry, we have many windows open. We have our practice management system. Our intraoral scanner may have another window. We may have a texting app with the patients. That is another window.

We have our imaging sensor. That's another window. So anytime we can reduce the windows, it is very, very valuable. And especially when you're talking about change management.

Right? So you're bringing in a new technology. You're bringing in AI, which even though it's it's seamless and it's simple and it's plug and play, it's still something new for your team members. Right?

And you can always get that pushback.

If you have one less window, meaning your team, if you're not on SOTA already, you should be. But if you're in SOTA, they're already interacting with SOTA. They're coming in, they're taking x rays, you know, they're doing everything that you would normally do in SOTA. But now you have the incorporation of AI within your exams already. So what's cool about SOTA is that they're really a true, web based imaging software.

So, it's not like you take the images on an app and then you later get uploaded to the web. No, everything is done in the web.

So, really cool. But let's say we go in and we look at some exams.

And I go in and look at a recent patient.

Go to Irving here. And I can just go into his Pano and ByteWings that was recently taken. And what you'll see here is that you see the little color at the bottom. So, green is for calculus, pink is for caries, purple is for notable margins. So, you'll kind of get used to this as you see it.

You can go in and, you know, just click on an image, and what you'll see is caries pop up. I mean, you don't have to be a dentist to see that something is different here, right? And your patient may even ask you, hey, doc, what's, what's the pink stuff?

So very nice to allow you to have that intelligent communication with the patients.

And what you'll see is when you roll over this, it will show you the, the breakdown of the caries. So this specific lesion is nineteen percent in dentin, eighty one percent in enamel.

And so if you're going and trying to explain this to a patient and you say, you see, missus Jones, right here, we have a little bit of decay, and they're going, what? What are you pointing at? I mean, we're so used to looking at radiographs that it pops out at us. Oh, yeah. Interproximally, mesial of the second premolar. For a patient, this is very difficult to see.

Now you go in within your SOTA imaging that you're already living in and you click on your AI viewer and you say, you see Mrs. Jones where this little pink blob is?

That's some decay.

Now why we need to treat and plan this and maybe even get it done today is this reason. And you turn on the tooth parts. And the tooth parts makes this light up, and the patient can really start to understand. Because what you can say is things like this.

You see how this little pink blob has gone through the purple and into the green area. The purple is your enamel. That is the hard, you know, protective outer layer of your tooth. However, this has gone through the enamel and into what we call the dentin.

The dentin is a soft, spongy area of your tooth. And, because of that, you know, we should probably get this treatment planned today.

How different is that case presentation? I mean, you can see why patients, the case presentation is going up by thirty percent even.

Because when this is explained well, patients really get it. Now, in addition, let's say you're looking at calculus. When you're thinking about scaling and root planing, you're talking to patients about calculus and whatnot, even looking at this, look at how difficult these little calculus ledges are. Right? They're there, but they're not easy to see.

And so when you go in and look at this and say, oh, you see these little boxes, that means you have some tartar. And that's why we have to do scaling root planing, for example. And let me just look at, let's look at another image.

So here again, you're going to see a lot of different detections. You're going to see interproximal. So this is a great example, right? We've got some bone level measurements that are different colors. So we've got a little bit of bone loss happening. We've got some interproximal decay.

We've got a potential little notable margin that we may want to look at. We've got calculus. But like when you're explaining to this patient why you may not need to treat these and you may need to treat this, it becomes so easy when you add in tooth parts. Because you can say, you see how this is in the purple. It hasn't gone into the green.

But how this one has extended into the green area. That's why we need to do that one. We may still need to do these lower ones at some point. Right now I'm going to watch them or I'm going to put you on a floor at rinse or I'm going to remineralize them, you know, with MI pace or whatever you want to do.

Having that intelligent conversation with your patients.

Now, in addition, when you're coming into your to your exams, it's very easy to to look at this and point your eyes to areas of interest and know, Okay, I got to look at this one. I got to look at that one. I want look at approximately one check the contact there. So very easy. Makes you very efficient in your exams as well.

Let's go back and we'll we'll kind of finish it off for today.

One thing that I am really passionate about with AI is that dentistry will surpass medicine with health care AI. You've probably seen the articles about health care AI.

There was one recently where AI was finding breast cancer even years before a human could find it.

You have it looking at brain scans, and even during COVID, they were using it for chest radiographs. So medicine really started out with the FDA approvals and getting it into hospitals. However, think of how often a patient is going to see that. Let's hope that none of us have breast cancer or brain tumors or something in our chest. But even if we did have that, we probably wouldn't see it. It'd be seen by a radiologist in a back room and it would never be seen.

But now with healthcare AI in dentistry, our patients see us every six months, every year hopefully, and we're showing them the radiographs. We're getting radiographs done every year, for example, and they're seeing it. For most consumers, dentistry will be the first time they see healthcare AI. And that I think is really cool. And that could be in your practice.

And think about how this portrays you as this first mover advantage be known in your community as someone that's using AI technology. I mean, we have stories of clients that are using this and they show it to someone and, you know, the woman says, wow, that's so cool. You think I could bring my husband in and we can take radiographs of him? Or they probably say X rays. Take X rays of him and look at what his radiographs look like. I mean, a lot of our clients are actually getting new patients because of this.

So I think the incorporation with what we're doing with SODA is so valuable because if you're already using SODA, very easy. I mean, plug and play to turn this on.

And if you're not using SODA, you can start with an imaging software like SODA that is fully on the cloud. You know, this is very important, especially if you're planning on scaling your practice or if you want to look at things remotely, you don't want to depend on servers.

And now just flip a switch and bring on artificial intelligence into your practice and have your practice really be a leader in technology.

So we've talked a little bit about inconsistency and inefficiency, the untapped data and the unexplored business opportunities. And I hope that you can see how by adding artificial intelligence, especially in your radiographs, you can get consistency and efficient decisions. You can benefit from your data and maximize your business opportunities.

So, if you're interested in anything that I do, here's my Instagram or anything that I do. I'd be happy to look through some questions and answer some questions.

You can put it on, the Q and A and I will go through and look at these questions.

And I'm sorry, if you put a question in before, it went away because we lost that last part of the webinar.

It recorded, so I'm sure we have it.

So, there's questions about cost, which I am not the right person to answer.

But I think if you are registered for this, we can reach out to you and we can have, the SOTA team or the Pearl team talk to you about that.

Which imaging software do you integrate with? Can you upload images into the system?

So, from a PERTOL side, we integrate with most imaging softwares. But specifically with SOTA, we have a native integration, so where you don't have to have a separate window. In any type of integration, you never have to, upload images into the software. It automatically takes them. So it's not like an extra step. They just show up with the AI already attached to them.

Can we get the recording of the webinar? Yes. It'll probably be in two parts, so maybe we'll edit it together and then send it out. But we should have have the recording of this and we can we can send this out yet.

And let me see if there's any other ones.

Is soda related to Pearl or a separate approach or product? So Pearl is within soda now.

So if you have soda already, you can turn Pearl on and it'll be native within that same window.

Yeah. Now, Perilla is also a separate product.

But, you know, for this webinar, we're talking about the, the SOTA integration. Yeah.

And any other questions? Okay, there's more coming in.

Do you have product that analyzes full CBCT scans? That will be coming probably next year. Yeah, so we've been working on that for many years, but not fully ready yet. Good question.

Yeah, so right now, in the U. S. We're looking at, bitewings and periapicals. If you're outside of the U.

S, you can contact us And, in certain areas of the world, we do have approval for panoramics as well.

So, yeah, I just answered some of that.

Okay, so any other questions, feel free to message us. Let me put my email back up here.

Okay, so there's my email, very easy, kylehelloperl dot com.

And if you have any questions, can put it there. But sorry about the technical difficulties today.

We will get you guys the recording and let me see if there's anyone. It looks like there's maybe some more chats coming through.

Great.

Okay. Well, thanks everybody for coming and sorry about the technical difficulties. Have a good day.

‍

‍