I'm gonna jump in because I know doctor Oster's got a lot to share with us this evening.

I am Sheela Roth, head of clinical education at Pearl. I am joined tonight by our director of partner growth, Kimberly Quieschen. She'll be here at the end as well. And I'm very excited to be cohosting this webinar with our partners from Patterson. So thankful to them as well.

Tonight, we're gonna dive into some very advancements that are shaping dentistry today from AI driven diagnostics as well as moving into chatbots, automated perio charting. So much in the industry is changing to help us take even better care of our patients and ourselves.

Tonight, we're gonna explore how these tools are transforming the way our practices are operating, saving time, boosting efficiency, and improving clinical outcomes, and overall helping with patient engagement and retention.

Probably by tonight, the end, you will have these little golden nuggets that you'll take back to your practice that'll AI future proof your practice.

And how are we going to integrate these new ideas is something doctor Osterwal will walk us through.

I'd love to know one more question before we dive in tonight. If you can put in the chat, how many of you are already using AI or automation in your practice today? We'd love to hear from you. We'll incorporate some of that in our q and a.

Pop your answers into there, and we'll get started.

Without further ado, I would like to introduce you to one of my favorite speakers, doctor Peter Oster.

He is a leading expert in cosmetic and reconstructive dentistry, practicing in practicing in Pomona, New York, a two time director of the American Academy of Cosmetic Dentistry, known for his passion, which you'll see tonight, for innovation and education, having taught dentists nationwide.

He's also been honored as the AACD Humanitarian of the Year and top dentist of Hudson Valley for over a decade. That's just a few things about doctor Oster. I'm lecturing all over the place. I'm thankful that he chose to join us tonight.

Without further ado, doctor Oster, I'll turn it over to you. And everybody stay tuned for an exciting evening. We look forward to seeing you on the other side for the q and a.

Thank you so much, and happy Thursday.

Well, I guess it's kind of a good day of the week. Thursday is toward the end of the week, so that's a beautiful thing. And thank you all for being here, listening to me today. I am very excited about this lecture. There are a couple of lectures I give that I'm much more excited about than others, and this is definitely one of my favorites because the subject is new and exciting and wonderful.

And I hope you're enjoying yourself, whether this is evening, afternoon, or for people listening to it tomorrow morning.

Whatever it is, I hope you enjoy every second, and I'm gonna do my best to make sure that you do. So what's hot in dentistry now? That's as good a title as I'm gonna come up with. So AI is here now, and it's here in dentistry, and it's taking over in many wonderful ways, and it'll rock your dental world.

Is it perfect? No. Is it great? In many ways, absolutely.

And this is something that I learned in dentistry maybe fifteen years ago.

Read it.

So you don't miss out on things if you don't try them.

You do miss out on things.

May maybe I didn't word that right. Let's try that again. You miss out on things if you don't try them.

You don't miss out if you do try.

Can you fail if you try? Absolutely.

But you can't succeed, and you can't learn new things unless you start to work on them. And that made an incredible difference in my practice and an incredible difference in my life.

So try new things, and it may be an incredible thing for you, your life, and your practice.

So the future of AI is now.

And much of it, we've already been using, maybe for many years. So the Florida Probe VoiceWorks and similar systems are basically AI.

Crown design using Exocad and other companies like it where crowns and bridges can be designed in provisionals and partials and smile design, that is basically AI that's already been around.

High quality scanners, which do amazing things. We'll talk about some new ones later. And they help with the software, they outsource your design, and they automate your design. So they help you use it in practice and outside the practice. But dental AI is far more than that already.

AI is proposed to reduce diagnostic errors by up to fifty percent in ten years.

That's incredible, and I'm clearly seeing it in my practice already.

The changes are amazing by using AI systems.

But in this world, will we be able to diagnose decay and fight illness before it begins?

And if so, when?

Now.

It's already happening. It's already changing what we can do in dentistry.

So if you look at this AI here, you'll see the things that have happened in dentistry over the last two hundred years.

Who knew that braces started in eighteen ten? I'm not so sure if I want to use the braces that were used in eighteen ten. I don't know.

But the high speed drill was in the nineteen fifties.

Implants well, really good working implants started in the nineteen seventies.

Digital dentistry is relatively new, but AI as we're starting to use it is brand new. And it's changing everything, and AI will probably change dentistry more than anything we've dealt with over the years. From fluoride to implants to digital dentistry to dental drills, maybe not to toothpaste. I'm not so sure about that. But certainly, the way our dentistry is going to go in the future. And it's happening incredibly fast. It's mind boggling.

So I'm just gonna cover there's actually something on my screen here that's stopping me from being able to see something. There you go. Getting it out of the way.

Oh, now now it's still there. Anyway alright. So it's exciting. It's amazing. And in some cases, it can be a little bit funny.

So what's the funniest way to start a lecture about AI?

Well, I'll read this.

Ladies and gentlemen, brace yourself for a riveting journey into the future of dentistry, where even the tooth fairy might be replaced by an AI powered dental assistant. So if you're ready to sink your teeth into the fascinating world of dental AI, let's floss our way through this lecture and see if we can extract some wisdom, teeth included.

Well, the sense of humor of ChatGPT may need a little bit of an improvement, but that actually is AI of a cute way. And this is the picture that it generated when I asked about it. Look at that poor tooth that I'm not sure what that tooth is. I don't know if the tooth has been extracted or whether the tooth is gonna have a tooth extracted. Whatever it is, it's very cute, and it was designed for what I asked, ChetGPT.

But AI is far more than chatbots.

There's so much more to it, and some of it is clearly amazing, and it's clearly happening now.

It's about her AI arm.

Alright. You've seen that before.

We've seen people who have artificial arms.

What we haven't seen are artificial arms that completely mimic every single movement that the other arm can do.

And that is a wonderful, incredible, amazing thing. So look what this woman can do. She can fix her hair.

She can cut an onion.

She can hold an onion.

That right hand can do virtually everything the left hand can because all the muscles of every finger are working the same way. The muscles around the elbow are working the same, and the shoulder is working the same. This is what AI can do in a wonderful way. And look at this. So this is a study that was in The Lancet, and it's mind boggling. So if you read about it, it says it revealed how clinicians had developed a deep learning model to identify which at risk infants have retinopathy of prematurity.

What that basically means is it can identify children that may be born AI, and they may be able to be treated in utero.

And doctors can't necessarily see it, but AI can.

The symptoms can't be seen by the naked eye. So the tool was trained using thousands of images of the eyes of newborns so that they can be as effective as senior pediatric ophthalmologists in discriminating normal retinal images and catch it before the baby is born.

Pretty amazing.

And then there's this, and you may have heard of this already.

AI is always be is already being used to help detect breast cancer, and they can improve detection by fifty percent.

So this is a story, and I actually saw a patient yesterday who's a physician.

And he said he will never send a patient directly for mammography.

He will always do a full examination on the patient.

He will always consider AI as well. And here's the reason.

So this is a case of a woman where seven out of fourteen diagnostic people detected cancer.

Then AI went into the picture. And if you look at the AI figure here, you're seeing a tumor that wasn't necessarily noticed by seven out of the fourteen diagnosticians.

Then after they saw this with AI, fourteen of the fourteen people agreed with the AI assistants.

How incredible is that? How incredible is it that we can save lives, that we can save vision, that we can save the ability to use an arm? These are things that AI can use outside of dentistry.

Of course, in our own field, we can find different things that it will do. For the growing number of seniors, it's gigantic.

Why? Have you tried to get an appointment with a specialist, a a medical specialist lately?

I found that it can be three months away, four months away. But now with AI, there are tools out there, including the Apple Watch, including other watches like it, other devices like it. And believe it or not, coming soon, they're going to be earbuds that will monitor blood pressures, will monitor murmurs, irregular patterns, arrhythmias.

All these things are part of dental telemedicine.

And say it's a senior who has real problems getting places.

They can be treated online. They can be medicated. The medication can be changed. All of these things are done with AI. So AI is amazing.

And look at this story.

So one of the first times that I really started to hear much about computers and what they can do in AI is when I saw The Imitation Game. If you haven't seen it, it's a really, really incredible movie.

But the other thing about it is what it showed, and it showed how we broke the Nazi code during World War two.

So this is Alan Turing's computer. Alan Turing was the guy we just saw. Actually, that's not him.

This is the star, Benedict Cumberbatch.

But what Alan Turing did was had this massive computer. Look at the size of this thing. This is what was used in the nineteen forties to break the Nazi code. So AI was actually present many, many years ago.

Then this is mission control in Houston in the nineteen sixties.

Take a look at the computers, the number of computers that were there.

Pretty amazing. And you would imagine that these things were incredibly powerful because they help people go in into space, and they were altered somewhat to send people to the moon.

But this is where we are with our computers and AI in twenty twenty five. Take a look at that. I'm sure you recognize that it's an AI, but it doesn't have to be an iPhone. It can be a Google phone. It can be an Android phone. All of the phones out there, the sophisticated phones, can do incredible things that are either AI related or pure AI.

And here's what's amazing.

This little phone has more information on it than this and this.

So those gigantic computer systems don't have anywhere near the amount of information that we have in our AI, but even more amazingly than we have in our watches.

So that little watch has more information in it than Turing's computer, than the Houston Rocketspace computer.

That is how fast everything is changing, and our world of dentistry and medicine and AI are changing.

So what is AI?

Basically, we're talking about computer systems that can stimulate the problem solving and decision making capabilities of the human mind.

And the building blocks of AI is basically how AI is trained.

So it starts with input and a tremendous amount of volume of data fed into computer programs. So we feed the information in.

But then the computer itself takes over. And there are algorithms that analyze the data, they detect the patterns, and they learn.

And then comes the output where the AI in the computer generates insights, predictions, or decisions.

And they look to help us detect problems and find things that are not going exactly the way we think they should be. So here are the types of AI and their applications. So before we go into the dental applications, let's go into the applications of AI itself.

So there's machine learning, natural language processing, computer vision, and generative AI.

We know about many of these things, and many of these things we're using already. So what is machine learning? Well, basically, it's a subset of AI that that has computer systems learning and adapting without following explicit instructions.

So they're fed information, but then they basically take over themselves.

And, for example, fraud detection.

So what it looks for is anomalies anomalies in a database.

So they can see where there's insurance fraud they can go they can see. They can see where there are problems in a database that look far beyond what the norm is.

That's what we're looking for in AI, and there are things that we can't necessarily do on our own without spending hundreds of thousands of hours.

Then there's natural language processing, and that's the field of computer science that focuses on translating text and responding to spoken commands and summarizing text. Well, of course, all of us deal with that. And many of us, of course, we have Alexa, we have Siri, and with that are speech recognition and voice to text. So that is basically natural language processing.

And of course, the Florida Pro works that way.

Then there's computer vision.

So computer vision focuses on computers to help identify and understand objects and people in images. So now we have a little stuff for you to try. So pull out your phones, if you're not watching this on a phone, and let's start with this. So say we're gonna take a picture of a little mini plant.

What are we gonna learn from that, or what's the computer gonna learn from that?

Very little unless it already has information about plants. So let's be a little more specific. Let's look at this little apple.

AI did the computer learn when they saw a picture of an apple?

Well, they learned a little bit, but they didn't learn a lot because there are lots of different kinds of apples.

So there's a rotten apple on the left. There's the red apple in the middle. There's the apple with, what is that, a worm in it on the AI, and then there's a healthy green apple on the bottom. Now the computer's never gonna know that a green apple is the same as a red apple or it's related to a red apple unless we start giving it that kind of information, and it's never gonna know the bad apples.

And it's also never gonna know this one. I don't know. Computer takes a look at that. It's gonna say something different about that apple.

So from that, we can weed out bad apples, and that's what companies can do. So by putting all these kind of apples on a conveyor belt, they can knock the bad apples out. So if you go to your store and you see a bunch of really good apples, if you wonder why only good ones are there, well, AI, it depends on the grocery store, you will find that the bad apples are gone, and it's because the conveyor belt knocked out the bad ones.

So try this. Once you go to your phone and you go to the photo section, so you're gonna click on photos on the iPhone. I think there's something very similar in Android phones. Up on the top when you see the search bar, you're gonna put an animal in. So what kind of animal do you have at home? You know, if you have a dog, if you have a cat, do you have a horse? Put dog in and see what happens.

And you will see that even at this point in time, AI is present. Even if you don't have Apple intelligence or any of the fancy new stuff out there, it already has differentiated.

So if you go there, take a look at your phone, put in dog, put in cat, and what you're gonna see is you're gonna see a lot of photos of a cat or a dog, even if it's not yours. And if you have other pictures of cats and dogs, they're gonna show up in the picture. So let me show you what I did, and maybe you can do the same thing. So instead of putting dog in, put in animals.

And if you take a look, this this is a new phone that I had. And what are you looking at? Well, you're looking at eighteen photos of animals that I had at the time. I think it was a month after I got the phone. So here I am with my my, cousin's poodle.

And if you take a look down below it, you'll see a dinosaur that was at a holiday exhibit.

And down below, you're gonna see a bunch of llamas.

Now how does that computer know that all of those are animals?

It has the database to be able to figure that out. So what I did, since some of these pictures are of llamas and alpacas, I clicked on llamas.

And if you take a look, it's knocked out all the pictures of dogs, and all you see up there are things that are resembling llamas.

But if you take a look at it, you'll see it's not perfectly accurate because it doesn't know the difference between a llama and an alpaca.

Well, that's okay AI I don't know the difference between a llama and alpaca either. So let's skip that part. But that's the kind of stuff that you can do now in your computer. You can put anything into photos, and it'll analyze the photos.

You put in Christmas, it'll show Christmas photos. You put in Thanksgiving, it'll show Thanksgiving photos. It understands just by looking at pictures. That's what it can do.

So here's a question. And, unfortunately, nobody can answer me now, but you can tell me this later. And that's who uses ChatGPT or another chatbot on a regular basis in their practice.

Well, it's time to start using it. So let me show you what we can do with just ChatGPT.

So if you take a look here, there's the there's the screen, and we're gonna put in a question. So the question is AI an email designed for a dental practice to send to their patients to reactivate overdue patients.

Look how fast it comes up with an answer. It's absolutely mind boggling. It's maybe two seconds to write a really good letter. Now are you gonna use that letter as it is? Not necessarily.

But you'll go through it, weed out the stuff that doesn't make sense, and then go further. So if you wanna change it, if you've got a periodontal practice or you do a lot of perio and you wanna bring in perio patients, then it will do this for you. And immediately, it's going to go to the heart of a periodontal practice or the periodontal portion of a general practice.

This is pretty amazing.

So think about this. I've got other ideas for ways to use ChatGPT or similar programs. Number one is you wanna get rid of a patient. Now is it often you wanna do that? No. But do you occasionally?

Yes.

And you want a good letter put together that's brought about from potentially Internet sources as the best way to write it. Now are you going to send it that way? No. You're going to have someone who has a legal background go over it, but it's gonna cost you a whole lot less to have it generated and chat GPT than have a lawyer write the letter for you. So all of these things make a lot of sense for your dentistry.

So AI at work, what does that mean and where do we use it? Well, there's deep learning outside of dentistry. Now look at this stuff.

So Shazam. How many people have used Shazam before? I'm sure almost all of us have. If we haven't, at the end of this, we can we can play with Shazam together.

But Shazam figuring out a song is amazing. I mean, you and on on, on an Apple iPhone, if you use Shazam, you click on it in the back. It's a free app. You put it on, and it shows the song that you're that someone's listening to in the room that you're in and the artist, and suddenly a picture of the artist shows up.

That is mind boggling.

Advertisers finding relevant ads for you by your age, your interest, gender, and location. Is that a good thing?

Yeah. Sometimes.

A lot of AI, not.

Then there's Netflix determining a show to watch based on your past shows, and that's actually kinda cool. So there are a whole lot of shows on Netflix or on some of the other services, but if you happen to like a show about physicians, it's gonna show you ten more shows about physicians or other shows that are similar. So all of these things are ways to use deep learning outside of dentistry.

Some of them are a little scary, and this is AI of the one I one of the things I was talking about before. So this is a photo that my daughter sent me.

She had found a really cool thing for my grandson.

And if you take a look at it, it's a track.

That's got an anti gravity dinosaur car on it.

Okay. That's cool. I thought it was a good picture. Now this came to me in a text.

Okay?

Later that day, I was on Facebook. And you know how on Facebook you see ads every once in a while popping up? Well, this is the ad that I saw, and this is a dinosaur truck and a little boy playing with.

And this I don't I don't look at kid toys.

So clearly, it somehow found my text and turned it into a potential sale for another product.

Now to show you what a dummy I am, I actually bought the product, and you'll see why. It's very cool. It comes with eight little cars. These are the cars over here. The cars go in the dinosaur's mouth. The dinosaur eats up the dinosaurs, and guess what it does?

Well, exactly what four year old boys and five year old boys think are wonderful, it poops out the cars.

Is there anything better than that? But is there worse than getting these ads immediately after simply a text? Absolutely.

Alright. Let's be serious. So let's move on to deep learning in medicine, and then we're gonna move on to dentistry.

Identifying locations and types of tumors, and it's not just breast tumors, it's brain tumors, it's all kinds of other anomalies that it can identify in ways that in many cases we can't.

AI robot assisted surgery, which is pretty amazing and it's used in dentistry now with implants. There are a lot of really great AI robotics that are already available.

This to me is one of the most important things out there. Because as a dentist, how many times do we need to prescribe something, but the patient comes in with a list of forty medications that they're on? You know, usually, I I kind of a I kinda figure out the patient's age by how many prescriptions they're on. And it's true.

So who looks through this stuff? Well, theoretically, the pharmacist does. But do you think the pharmacist can go through every single of the forty prescriptions to make sure that what you're giving them isn't going to have a problem with it? No.

It's not. So this is a very, very cool thing that AI can do and, of course, in the early detection of illnesses as well as we mentioned before. So the advantages of AI are number one, speed. It's many times faster than anything we can do.

There are no preconceived ideas. The computer just works onward. And more important, it can still keep working at the same level and strengths for hours and hours and days and days. Make imagine what it'd be like for working five straight days on something twenty four seven.

Computers can do that. So they can go through thousands and thousands and thousands and millions of pieces of data that we can't possibly go through in a short period of time.

And there's diagnostic consistency that can come from AI.

Now there are disadvantages of AI too. Now number one, AI is emotionless. Computers don't have emotions.

They don't improve in many ways.

There are high production costs in anything we're doing with AI. They're responsible for future unemployment, and all you have to do is go to the supermarket these days and see how few cashiers are there anymore. And there's a lack of ethics and moral code of conduct. So there's certainly disadvantages as well, but there are so many more advantages in terms of AI. And in dentistry, we're basically dealing with inconsistency.

Digital dentistry's untapped potential and bottlenecks and inefficiencies that we can't find but the computers can.

So how do we do that? Well, here's a question for all of you. What kind of hygienist do you have in your practice?

Are there young dentists in the practice? Because there are two kinds of both.

There are hygienists that feel comfortable diagnosing, even though that's not theoretically what they're supposed to do, and there are hygienists that don't wanna touch diagnosis.

There are young dentists who think they can diagnose, and there are many young dentists who can't. And this is where AI is huge. For young inexperienced dentists, for uncomfortable hygienists, and for offices that analyze practice management data, And for DSOs and multiple office groups, they wanna compare and AI the data between one office and another so they can learn from one office and move the information over to another.

So AI and dentistry, already we're doing all of this. We're working on enhanced scheduling we'll be talking about today. Main thing we'll be talking about is diagnostic assistance, insurance verification, and workflow and staffing.

Much of what we'll be talking about today comes from a company called Pearl, which to me is one of the finest companies in dentistry, and it hasn't been around very long.

It started by a dentist, and it is a company that truly cares about creating something that is helpful for thousands and thousands and millions of dentists around the world with a huge database of information.

So there's dental AI that we're already using. Now for many of us who have, scanners, well, stitching is AI. It's taking different scans or different photos of the occlusals or the buckles or the linguals of teach teeth and stitching them together. And it's getting better and better.

So these are the latest scanners on the market. The Trios three shape, the Itera Lumina, and the DEXIS thirty eight hundred. And there are other new ones on the market too, which stitch better and better elimination of mouth items. They immediately will knock out artifact.

They'll immediately take the stitches and put them together. In the old days, we had to turn the computer off between stitches. We don't have to do that anymore. It just automatically takes one, latches it to the other, and you can keep doing until you're finished, and the scans are so much faster and so much more accurate.

So with all of that, why will Dental AI leapfrog the rest of medicine? Well, it's got a number of reasons. Number one, we take more X rays than anybody.

Anybody. I I mean, how many times do you go to your general physician and have an X-ray taken? Only if you got, you know, potentials for for pneumonia. It's rare that they ever take X rays unless there's major symptoms going on. There are lower stakes in dentistry because there's not a fear of that way AI can affect somebody's life as much. There's clinical inconsistencies.

And in order to give patient trust, we'd like to get our dentistry a little more consistent. And there are far fewer radiologists in dentistry than there are in medicine.

So as far as radiology is concerned, there are one point four billion x rays captured annually in the US.

Thirty nine percent of medical X rays are dental. How crazy is that? So we take almost all of the X rays in medicine.

Eighty six percent of US dentists, have dental image cap or captured, and only ten to fifteen percent do in other fields of medicine.

That's why this is going to skyrocket in dentistry.

So what's our goal in using AI in dentistry?

It's to assist us. It's not to make our decisions. It's to help us and guide us through decisions, but we're the final decision maker. And they're basically looking for items that deviate from the norm.

We wanna go up against clinical inaccuracy.

So take a look at this. This was a study done at UCLA.

Dental students, if you look at the sensitivity, specificity, and accuracy of x-ray reading, they were in the fifty percent range.

Once they added AI to it, they reached the ninety to ninety six percent change. But that's just not dental students.

That's dentists in general. Years ago, there was an article in a magazine called Reader's Digest. Some of us older people here will remember that. That was an article that destroyed dentists, which was odd because Reader's Digest didn't doctor any field except dentistry. It was very weird. But the article showed a whole bunch of people giving completely different AI, and that became a real problem for a period of time for dentistry.

And that happens all the time. That is where AI can really help us.

So just to go over this, I usually do this with people in a room so it's a little bit more fun, but this is a story of a hundred and thirty six people looking at the same x rays.

K? And this was the full mouth series that they saw.

Here's what they came out with from this FMS. Let's just take this picture. So what tooth do you think we're looking at in particular? Let's look at tooth number three.

And if you look at tooth number three, what do you see? Well, you may see some recurrent decay up there.

You may see a fracture up there. You may see more decay higher up. You may see some decay over on the distal side. Not exactly sure what we're seeing. We're only looking at an x-ray, but many people, a hundred and thirty six, looked at the same x-ray. Let's see what they found.

Fifty one percent oh, my this the screen is not showing the way I want it to, but you'll get the idea.

Fifty one percent saw no caries.

Forty eight percent saw caries.

What made this even weirder is thirty nine percent saw no recurrent decay, and sixty point three percent showed recurrent decay. Now why is the number on the top of the page different than the number on the bottom of the page in either of them? None of this makes a whole lot of sense to me because dec recurrent decay is decay.

But here is an example of the discrepancies of what people see when they see an x-ray, even one that looks reasonably simple.

Let's take a look at another one. Okay? So now we're looking at tooth number nineteen, and we're seeing different kinds of things here. So I'll I'll tell you the things that people see when they look at these X rays. So they see potential recurrent decay or new decay up there. They see something going on with the pulp chamber on the distal side where it's kind of shrinking.

So what are they looking at here? Is this an endodontic problem? Well, maybe. Doesn't look like the decay is all the way to the pulp, but you can't really tell completely without the PA.

But patient the people were told, the dentists were told to say what this this case would cost in their office to fix. So let's see what they said about number nineteen.

Twenty six percent said it would cost zero to five hundred dollars. So assuming that's a filling, a large distal DO filling or a DOL filling, something like that.

Thirty seven percent said five hundred to a thousand dollars.

That's kind of a tough question. I'm not sure what five hundred to a thousand dollars is because it's too little for a root canal.

It's too much for a filling.

Maybe it's an onlay, but I don't know how many people actually are doing onlays. The thirty seven percent would say that. So let's look at the next batch. Twenty one percent said one thousand to three thousand dollars.

What do you think?

Take a look at it yourself. I'll show you that x that X-ray one more time and tell me what you saw.

There you go. There it is.

So how much money would that cost in your office? I mean, it's very different. Would you be in this group? Well, that group is three thousand to five thousand dollars. Now I'm assuming that that's talking about a crown, a post, an endo, maybe crown lengthening, or maybe an extraction and an implant.

Whatever it is, there's a very big difference between that sixteen percent and the twenty six percent.

That is how much discrepancy there is in people reading the same x rays. So of those a hundred and thirty six people that saw those x rays, the treatment plan for the entire mouth was between three hundred and thirty six thousand dollars.

If that bananas, that's what we're dealing with in terms of the difference between what different people see.

And a failure to diagnose claims can be catastrophic.

So if you take a look at this, what do you think this is?

It's an ad.

It's an ad that I found from a lawyer that says, see if you have a case for dental negligence.

And these are the things that can that they can win money for, failure to diagnose gum disease, periodontal disease, oral cancer, serious tooth decay, root canal treatment.

It's important to be able to read all x rays in a good way.

And in fact, from studies we've seen, forty three percent of caries and x rays are undiagnosed.

Twenty percent of caries diagnosed are not actual caries, and forty nine percent of radiolucencies aren't diagnosed.

That is a real problem because those periapical radiolucencies can be disastrous.

Second opinion is something that Kearl has, which is mind boggling.

It's the world most comprehensive FDA cleared real time AI diagnostic aid.

Take a look at that X-ray.

How much better is it for a patient to see that than to see a typical gray, black, and white x-ray?

It changes everything in terms of what they see, what they read, what they understand.

And these are the kind of things, these are many things that Pearl can show you. They can show you periapical radiolucencies, but you can see them in a color, in blue, so much easier for a patient to understand.

It can show you the color of margins.

You can see where there are periodontal defects, where the orange line in the middle there is larger than the white line, which is more normal.

You can see root canal issues.

You can see problems on panorexes.

You can see calculus, very simple calculus that's very hard for patients to see. Look at the impaction in that right side. See, each of them has a different color.

Caries, the way that Pearl does it is amazing. So it'll tell you what's cementum, what's dentin, and what's enamel, which is helpful for you and it's helpful for the patient. Now if you take a look at the x-ray on the right side, what do you see? Well, you're basically seeing a lot of decay that's only in enamel.

Once that decay is beyond the enamel, what do we do? We treat it. But if it's only an enamel, we don't necessarily treat it. This is something for a patient to truly understand why it's important for them to floss every day and maybe never turn those things into teeth that actually have to be treated.

So what are we looking at? Well, I like to call this Fifty Shades of Grey.

And it has nothing to do with the book and nothing to do with the movie. What it has to do is the fact that the human eye can see up to Fifty Shades of Grey, whereas computers can read up to six to eight hundred shades of gray. So it can be far more accurate at diagnosing things in many ways that us humans can. So if you look at this screen over here, this is not the full dental version of it that patients can see, but it shows us what's happening on a tooth. So if you look up here, you can see a tooth where this is mostly I believe this is all in, enamel on the screen. If you look at this screen over here, obviously some of this is enamel, some of this is dentin, and this is work that has to be treated. But if you look over here, the lighter pink is only in the enamel and does not have to be treated.

So in this, if you look at it further, it's twenty seven percent in enamel, seventy three percent in dental in dentin rather, so that tooth definitely needs to be treated.

So we can see the progression of cavities. We can see, again, on the right side, the only in enamel pink, and on the left side, kind of a circular view of where the caries has gone into the dentin as well. If you're looking at this screen over here, you can see calculus.

And over here, you can really see a view that we can understand and a patient can understand about periapical radiolucency.

So here's a screen that's got a bunch of things on it. So if you take a look at it, you can see that there are, on the top tooth, you can see only problems in enamel.

On the bottom, you can see some calculus.

You can see areas that are incipient, and this will tell you that it's a hundred percent in only in enamel. If you see here, it's something very different under that old filling.

So it's night and day for a patient to be able to look at it in this gray or to look at it in this color where you can tell patients this is the pulp chamber, this is the dentin, this is the enamel, this is caries. On this, it's very clear where it's a little darker red, so you can see it's actually hit a little bit of the dentin. In that case, it has to be treated. In this case, it's only in enamel, so it doesn't.

So what can this do? And this is amazing. This is a printout that you can look at every single morning at the morning huddle. It's got scheduled treatment, it has unscheduled treatment, and it has predicted undiagnosed AI treatment.

So, basically, it goes over a database of x rays. I believe it's eighteen months of x rays. It can go back into the system and it will tell you areas that may not have been diagnosed, may not have been noted before.

So these are patients who are already coming through the door that day, and if you have openings in your schedule that day, voila, we filled an extra hour. We filled an extra two hours. We could also tell from that unscheduled treatment and remind the patient that this is work that can be done potentially that day at the practice.

Pre check is a next generation thing that Pearl has, which is really amazing. So it's insurance eligibility and benefits verification, and it's done already powered by AI. So what it is is if you can find something the patient needs and it can be something that was diagnosed that day or something's diagnosed in the past, Immediately, it'll go right to your insurance company and determine what the insurance company will cover for that procedure that day.

So not only have you found the treatment to be done, but you've also found out what's going to be covered for the patient's sake. Coming soon.

Three d CVCT.

This is the future where you can do this on a CT scan.

It's getting amazing. It could be used with root canals, maxillaries, sinus information, the entire pulp system, orthodontic case tracking. The future at Pearl is absolutely amazing.

If you have questions, just a couple of minutes, you can ask them.

And the question is this, should we be diving head first, or should we tiptoe into AI and into other parts of dentistry? I I think it's time to dive in because many others are. Insurance companies are diving in. I'll tell you something about Pearl. Pearl was established by dentists.

There are other companies that were established for insurance companies to find treatment, not to treat.

You're looking at something which is going to help us decide what really, legitimately should be treated.

So be selective.

Give AI a try.

It can save you valuable time and valuable money because you can start to fill your schedule immediately that day.

It's not the answer, but it'll help you get to the answer.

So how do you end the lecture like this? Well, what's the most humorous way to end a seminar about AI?

Well, chap EPT said, before we wrap up, just a friendly AI, when the robots inevitably take over, be sure to compliment their algorithm before asking for mercy. Thank you all for attending. And remember, if you can't beat AI, at least try to make it laugh. Good luck out there, humans.

That's a little scary before the robots inevitably take over.

And I'd like to thank my good friends over at Pearl, and Mike Buckter has been my good friend for years. Sheela Roth is my good friend for months. And I have a new friend, and her name is Kimberly, and her last name is pronounced Quishin, I think. I hope I get this right. I also would like to thank Patterson Dental that has actually helped us with this seminar, and I thank you all for sticking with it. Thank you.

Amazing, doctor Oster. Thank you so much. I'm sure if we could give you a round of applause, it would be a standing ovation.

Such an incredible, you know, way of looking at where dentistry is going. And those of us that have been here a while, all of these things are extremely exciting. I know Kimberly has some questions that are very insightful questions about, do we jump into AI? Is there a learning curve there? Is it is it concerning? So, Kimberly, can you kinda lean into some of the questions that the audience had for us?

Yeah. Perfect. And we It was question. Right?

You nailed it. Yeah. Thank you, doctor.

It was perfect. Perfect.

Yeah. No. There are some great questions. So we wanna get to all of them.

Question about actually, Sheela, you mentioned it. What's the learning curve for these types of AI tools as practices are starting to put them into practice?

What's the learning curve?

It's it's not as bad as you would think it would be. Let's put let's put it that way because it looks like it would be really scary, and it's really not. And and I started implementing it, I'd say I'd say a week later, after I got it, and and implementing it pretty well. Now now one of the important things here that I really didn't mention is the quality of the X-ray.

And the quality of the x-ray is everything here. So if if anything else comes of this, it should be staff members learning how to take better, clearer x rays with less overlap because those x rays are gonna be great, and that makes your life a whole

lot easier.

Once you see the pictures up there, it becomes easier for you to understand too. It's not only understandable for the patients, but it's also much more understandable by you if you can see it so clearly.

Now what I wanna explain to you is second opinion works with a lot of imaging systems, in a standalone product. So you just open up second opinion dot com. You have your images there.

My favorite is also though when we move into our native integrations AI Eaglesoft and Open Dental and many of these where Pearl will live inside of your imaging system. So you click a button, and you're already looking at your Eaglesoft Imaging or Open Dental, lots of others, and the images will appear with the detections inside of your system.

We also know, like doctor Oster was saying, that we are built by clinicians for clinicians. So although the AI AI defaulted on typically, the click of one button or one space bar, you can default it off, look at the images, and be able to make your diagnosis as we are the second opinion, the backup to the doctor, and not diagnosing for your teams. Right? We're just highlighting pathologies as the doctor remains the the one in the driver's seat.

I have two things Perfect. I have two things to say about that.

First of all, as far as Canadians are concerned, I lectured about about AI in, at the Toronto meeting and also the Montreal meeting last year. And so I got probably the most and the best questions from the Canadians that I got from anybody. Yeah. They're they're just awesome. I I I wanna move.

No. I I I loved it there, and I love the people. The other thing I wanna say is is using using this in your practice, it's not only the fact that it makes it easier for the patient to understand it.

It's also the excitement when they see something completely new. And everybody sometimes I'll just put a picture up on screen, and patients without seeing anything about AI, and they'll say, what's that? You know, it's like it's like years ago the first time we ever used digital X rays as opposed to those little plastic things you put in people's mouths, and and people are really excited by it. So that's and anything new excites a patient.

Love that. Okay. Now these next two, I'm gonna combine into one, because there's general questions about accuracy, specifically clinical attachment levels, radiologic, bone loss. But then also there's a question about flexibility.

If you feel like something's been missed, can you add it? What are the options there as well? So sort of those two go together. Accuracy and what do we do if we disagree with what we see.

It happens.

It it happens. I mean, I'll take you know, if we're taking a full mouth series, sometimes you'll see something that looks like a problem on one X-ray of the of a tooth, and then on the other X-ray of the tooth that looks fine.

So you you know, this is your judgment. It's not it's not the judgment of the company. It's the judgment of the dentist or the hygienist that's looking at it. And and yeah. Yeah. I mean, I'm I'm you know, of course, there are going to be errors. There are gonna be errors that dentists make looking at their own X rays as as well.

But that's our job. Our job is to is to take the information that we've got and run with what information we're given.

Yeah. And I would also add to that, we spoke at the beginning of the q and a about the quality of the images. Right? And we all know, that AI is trained by humans.

Right? The more diagnostic the image is, the better we can review and make our decisions as clinicians. And that would be the same for the AI. So if we have, you know, elongation, foreshortening, you're gonna see false positives.

You're gonna see that AI lives in shades of gray, like doctor Oster was saying. If you're taking slanted images, your bone levels are going to be off. So the nice thing we've seen in regards to the bone level measurements is we are measuring that space between the AI finding that CEJ and that crest of the bone.

The the great thing here is very quickly, it gives your hygienist not full probing. We would love that, wouldn't we, hygienist? But it gives you a quick indicator of the areas I need to be mindful of today, right, which are those areas where the bone levels are yellow, giving you that warning sign. The nice thing is turning that to the patient, they're now informed. These three areas back here, Kimberly, I know it's not Kimberly, we need to have you flossing a little bit more. And I would say cherry on top is taking a screenshot of those images and submitting to insurance will really help provide one more layer of documentation to decrease those scaling and route planning denials.

And I think that's the the other key win here is being able to say, we're gonna give all the documentation, and now you have one extra thing to just submit in there with your bone level I mean, sorry, with your probing depths and with your radiographs, I think that's one more thing you your team can count on because it's really important. Not to mention boxes of calculus are right there for your patients to now see and understand the scaling and root planning needs.

Well, I think between CT scans and the AI, we have so much more information than we had three years ago or four years ago. I mean, it's mind boggling. And and I I've gotta say, you know, are are there things where you're where where maybe there's a false positive?

Yes. But there are many cases already where I couldn't see anything on an X-ray. I just didn't see it.

But the AI showed it to me, and I did it. And suddenly, you're taking a look and you see this white black decay that you never knew that you wouldn't expect because it's not on the occlusal surface. It's just, like, three, four millimeters lower. So it it really has has helped us gain access to things we never would have seen before.

That's a great point. And, also, when you see this image that we have up on my screen here and those of you that are looking at, like, tooth number twenty eight distal, you can see that Pearl purposely colored this in a light pink area.

That lesion that we are, recommending the doctor take a look at is showing you that this area is still within that enamel.

Now you may say, why is Pearl showing me that? I'm never gonna fill that for two years. We hope you don't fill that if it doesn't need to be filled. It's a great way for you to talk to your patients about, we've gotta take care of tooth number thirty.

That's obviously why you came in. Let's keep an eye on the couple teeth in the front as well. Maybe need to be flossing a little bit more. Oftentimes, you may not wanna have that area on your radiograph.

We can simply hover over, click it off, and it'll disappear.

If you see something that we missed, which is gonna happen probably a little less frequently, you're going to be able to draw a cavity in if you prefer. Right? And it's they're gonna know that that's done by you. That's the point.

AI is not meant to be perfect, but when it comes to being that second set of eyes, especially when we look at those statistics of forty nine percent of periapical radiolucencies going undiagnosed, we're thankful that AI is there just like we're thankful that we have, you know, spell check. Right? I'm I'm really thankful it's there, but it's not always right. And I think that's the important part is to understand the expectations and to know that it's not a diagnostic tool to tell you what to do.

It's just highlighting pathologies for you to take a closer look at.

Isn't that the case? How many times have I tried to write good and it autocorrects to food, but I still am very happy that I have my autocorrect.

One question, and I know this will differ depending where where you live, so depending where this question came. But talk to me about panels and the ability for detections to appear on them.

So panels are going to be released soon.

Currently, they are not, released in the US. We do have them approved in the EU, and you do have the capabilities coming soon for caries, calculus, and periapical releasancies on panels. I think our exciting initiative doctor Oster talked about was the, release of the three d imaging to be able to initially just highlight some anatomy, being able to really kind of assist with measurements and implant placements.

But, again, by the end of the year, hopefully, some additional diagnostic aid in temporomandibular joint disorders, what are additional pathologies we need to be looking at when we when we start increasing our scope of diagnostics and having that second set of eyes point out certain specific flags of cysts or areas of concern that we need to take a closer look at.

Perfect. And panels are available in Canada as well. So just really everywhere excepting AI.

I kinda think that Canada is the US, honestly.

I know. I know. I know.

Careful.

I know.

I know. I regret that. We'll have to edit that out. Sorry, Canadians.

I apologize. That went to the left or to the right or upside down, but I love them all. That's my point. We're all in some respect.

Speaking of loving them all, we have, nailed all of the questions that everyone has sent in, so thank you for that.

Yeah. It's a great way to end.

Well, I would like to invite you to take some time with your team if you're interested in, getting a demo to sit down with your doctors and maybe even your whole team. Have them take a look at it. Is this a good match for your team? Right? Ultimately, is it a great match for your patients?

And for me as a hygienist, it's such a relief to be able to you know, when I used to check all my charts, who's coming in today, what unschedules dentistry is still remaining, we know that those areas are gonna get larger, and they're going to get more invasive, and they're going to get more expensive.

So if we have a tool that's now going to say, okay, Bob. We saw you last time. We recommended some fillings. Let's take a closer look at these today, and let me show them how far they're advancing.

Do you have any extra questions? Right? They're now able to see what you're seeing. The goal is not to use AI to diagnose more.

It's to use AI to have the patients understand and be informed to make their decisions so that they will accept that dentistry earlier and more often for you, Kimberly. Right?

Thank you AI for joining us this evening. We're really thankful you were here. Please stay tuned for additional webinars. We'll have doctor Oster back.

And, on our pro website, you have additional webinars coming up as well. And, again, feel free to, scan that code, grab a demo. Your Patterson team is standing by as well. Thank you again for joining us, and, have a great night.

Bye, everyone.

Bye, everybody.

‍