Dental AI & Patient Care

For a long time, dentistry has watched the AI revolution from a distance. While medicine has begun to harness AI’s potential in the realms of diagnostics, pharmacology, education and more, dentistry has been slower to adopt it. Until now.

With the pandemic of 2020 came a great push forward for telehealth—the practice of seeing patients remotely by video. Dentists comfortable working with 20th century tools suddenly had to rely on technology to survive. Now, earlier adopters find themselves thriving.
“Dentists were shut down overnight. And a lot of dentistry is continuing care,” said Dr. Cindy Roark, DMD, MS and SVP and Chief Clinical officer at Sage Dental Management. “The people who were successful were people who had already dipped their toe in the water and had teledentistry systems in play.”
Dr. Cindy Roark, DMD, MS and SVP, and Chief Clinical officer at Sage Dental Management
“Dentists were shut down overnight. And a lot of dentistry is continuing care,” said Dr. Cindy Roark, DMD, MS and SVP and Chief Clinical officer at Sage Dental Management. “The people who were successful were people who had already dipped their toe in the water and had teledentistry systems in play.”
- Dr. Cindy Roark, DMD, MS and SVP, and Chief Clinical officer at Sage Dental Management
Teledentistry is just the tip of the incisor when it comes to the application of new technologies in dentistry. “Dentistry is a manual skill. I can’t fix your tooth through a camera. I can’t even completely diagnose you through a camera. I need X-rays for that,” Roark said.

The real revolution will come when dentists begin to apply the AI in their offices. Why? Dentists struggle with an unrelenting workload that includes high patient volume—20 a day, according to the American Dental Association—as well as high administrative burdens and workflow inefficiencies.

The AI software available to dentists today helps them work smarter, faster and more accurately than ever before. And when working in concert with other technologies, AI will open opportunities beyond the four walls of a dental practice.

Here are three aspects of patient care that may be improved by AI integration.
01
Diagnostics
Dentistry has a trust problem, to put it mildly. Many adults are flat out afraid of going to the dentists. Others simply don’t trust their dentists to give honest and accurate recommendations. That mistrust is likely rooted in negative experiences.
“Dentistry has needed consistency for a very long time. I think we have all at some point seen studies that show that if you give 10 dentists a chart and x-rays, you will get 10 different diagnoses,” Dr. Roark told Authority magazine. “Diagnosis is highly variable, and variability leads to confusion and distrust from patients.”
Indeed, a recent study by the Dental AI Council asked 136 licensed dentists to provide diagnosis and treatment plans based on review of an FMX image set. Those dentists never delivered more than a than 50% diagnostic consensus on any given tooth and produced widely divergent full mouth treatment plans, with costs ranging from $300 to $36,000.

AI can help. Artificial intelligence has been supporting radiologists in medicine for several years, in part because of rapid advances in the field of computer vision and neural networks. In one case, an AI system built to distinguish between croissants and bear claws in order to automate checkout at a Japanese bakery was ultimately adapted to recognize sickle-shaped cancer cells.

It works like this: AI systems use computer vision to analyze images on a pixel-by-pixel basis, and incorporate neural networks that learn how to recognize different, relevant aspects of a diagnosis. In a network, the first layers of neurons (algorithms with their own sets of weights and biases) may distinguish an object’s general shape, while layers farther along in the network suss out other relevant details. In the case of recognizing cancer cells that might be the color of the nucleus, its size and roundness.

In dentistry, the same method is applied to learning the difference between caries—a cracked tooth versus a cavity versus gum disease. The AI then provides consistent diagnostic recommendations to dentists, who make the ultimate decision about diagnosis and treatment.
“The images come up annotated,” Roark said. ‘It’s showing the patient exactly what it's measuring: here's your bone loss, it's undeniable. Patients accept it tremendously well.. Because it’s like an objective third party.”
This improved consistency will improve trust in dentists over time. But studies have also shown that when human decisions are validated by machines, people are more likely to trust those decisions.

AI boosts confidence among patients.

63%
of patients are uncertain that their dentists are reading their X-rays correctly
55%
of patients would feel more confident if another dentist viewed their X-rays
82%
of patients would prefer that their dentist use the most advanced AI and diagnostic tools available.
90%
of dentists fears being sued by patients—British Dental Journal, 2018
02
Restoration
Dental restoration hasn’t had a great leap forward in about 15 years. While the advent of intraoral scans eliminated the need for crafting and shipping physical impressions, the process remains incremental. Submitted scans are often sent back for reimaging. When scans are usable, they still need to be marked by hand, introducing errors and often resulting in flawed finished fabrications that require remakes.

AI is creating a smoother, speedier process. Within five years, dentistry could achieve a 0% remake rate.
Here's How:
New scanning technology, powered by computer vision, provides dentists with real-time feedback. If a scan isn’t perfect, the system notifies the dentist immediately, allowing them to take another scan while the patient is still in the chair. AI margin marking means acceptable scans can pass seamlessly from the dentist's office to the laboratory design interface.
Generative algorithms – deep learning algorithms that create new images on their own– are applied to automatically design features that match with surrounding teeth.
Robotic milling machines manufacture the restoration to the design specifications and the quality of the final product can be reviewed by computer vision analysis to ensure that it perfectly matches the design.
These AI capabilities can be retrofitted to existing processes too. AI can also be integrated into chairside CAD/CAM systems that, themselves, have inspired much of the laboratory investment in AI-driven efficiencies. And even when physical molds are delivered to the lab, technicians can capture AI-enhanced scans to ensure accuracy.

Over time, data analytics systems will use digital data to track and identify longer-term trends, identifying the strengths and weaknesses of design methods, material choices and more.
03
Surgery
Dentistry is, at its most basic, a manual skill. So the idea that we would put the most delicate procedures in automated “hands” might feel counterintuitive. But robot surgeons have been with us for almost 20 years, allowing surgeons to control multiple robotic arms through a hand-operated console, giving them greater dexterity and vision when operating in hard-to-reach areas.
In hospitals, medical surgeons are already using robots to improve outcomes in coronary bypass surgeries, tumor excision, hip replacements, kidney transplants and more. It can be minimally invasive and puts the surgeon at the helm of a system that responds to its hand and finger movements with precision. And efforts are underway to fully automate some simple, repetitive maneuvers, like suturing, using the same AI that underpins self-driving cars.

Robot-assisted surgery aims to give dental procedures the same boost. Currently, the Yomi Robotic Dental System is the only FDA-approved system available. It combines AI-powered navigation software with haptic feedback—the use of touch to communicate information with users. Taken together, the Yomi helps dentists navigate the far corners of the mouth without worrying that their hands might falter.
Robot-assisted surgery is already being used to assist dentists with implant treatments, and could evolve to assist with endodontics, orthodontics and periodontics, always with a human dentist at the helm seeing the process through from start to finish.
“Trying to automate a process from soup to nuts, its just a lot harder than dividing the labor and finding places where the humans can play to their strengths, and the machines play to their strengths,” said Erik Brynjolfsson, director of the Digital Economy Lab at Stanford University.
For AI, that’s data analysis and predictive tools. For robots, its consistency.

So robots may do well with precision drilling or suturing, while dentists apply soft skills that machines can’t handle, like patient relations and care management. None of that is to say you’ll be seeing robots in your office soon.
“There’s one robotic implant placement system right now and it’s well over $100,000,”
Roark said. “So if the average dentist is not doing implants, a hurdle of that price tag is prohibitive.
‍
Affordable AI is more likely to become commonplace because it automates some part of a daily procedure that might be considered precision or low-skilled, Roark said. A robot that fills cavities, for example, or AI that generates a crown with perfect margins and a 3D printer that assembles it.
“I think when we get to the point where AI software can help automate that—golden, everyone will adopt, Roark said. “People don’t like spending money, unless it makes your life easier.”
Read Previous Post
Building ethical AI
03
Coming Next Month
Business Operation

Get Dental AI 101 in your inbox!

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Consent Preferences