Interview with the CEO of Pearl Which is Applying AI to Dentistry

Nextbigfuture interviewed Ophir Tanz, the CEO of Pearl. Pearl has been developing and using AI to improve dentistry for the past three years. The company officially launched last year.

Ophir was CEO of GumGum for 11 years. GumGum uses artificial intelligence and computer vision across media, advertising and sports. Ophir grew the company to hundreds of millions in revenue and hundreds of employees across four continents. Ophir was CEO and co-founder of Mojungle.com, a mobile-media sharing platform that was sold to Shozu.com in 2007. He co-founded and sold Fluidesign, an interactive and branding agency.

Pearl Dentist AI

There are over 115 startups applying AI for healthcare. 39 of those AI startups are applying computer vision and AI for radiology and other medical vision applications.

The reason there are so many companies working on vision and AI is that AI’s can be trained over millions of images while a human specialist would train with about ten thousand images. The increased volume of images enables the AI to become better than human specialists in identifying and anticipating disease from medical imaging.

Pearl is leading the way in using AI to help read dental X-rays instead of imaging of brains, kidneys, breasts, and other medical areas.

They can identify issues earlier and with higher precision. The Pearl AI recognizes dozens of dental pathologies, existing restorations, and natural anatomy.

Pearl has the largest collection of dental x-rays and has had thousands of dentists annotate the images.

Ophir Tanz told Nextbigfuture that a major problem in dentistry is the inconsistency in treatment. This can be the result of dentists getting overworked. They can see 20 or more patients in a day. The inconsistencies are not just in the treatment of dental patients but also with dental insurance claims and many other aspects of the dental industry.

Ophir said “AI can deliver comprehensive consistency to the dental industry”. Insurance companies are the financial foundation of dentistry. Hundreds of millions of claims are submitted every year. Only a few percent of the claims are reviewed. Millions of fraudulent or otherwise invalid claims slip through.

AI can quickly sanity check claims. Consistency can be checked by matching x-rays against the stated diagnosis and treatment. Only inconsistent claims will be subjected to manual review. The speed and efficiency of claims review increases, fraud, waste and abuse are reduced. Instant claims settlement will eventually become possible. Having lower insurance costs will make dental care more affordable and the quality of care will go up. More people will go the dentist if it is more affordable and this will help with the prevention of dental problems.

The time pressures in dentistry and medicine in general means that AI is needed to take the most recent observations and x-rays and tie them back to lengthy medical records. Pearl analyzes unstructured data within a patient’s medical record and ties it back to more massive datasets, then uses predictive analytics to surface other health concerns. It is not just helping to read the current x-rays but to look at changing gums from prior exams and identifying trends and changes.

AI and automation will also improve the precision in the creation of crowns.

The system helps dentists have an instant second opinion to ensure that possible problems are not missed. Dentists will also have more time to spend with patients and prescribe better treatment options. There is also a system that helps dental labs.

Having AI handle distractions has already been shown with driver assistance in cars. Tesla’s autopilot reduces the number of accidents from 1 in 500,000 miles to 1 in nearly 5 million miles. The autopilot takes away 20% of the driver focus on controlling speed and maintaining lanes to enable 100% of driver attention to look at the other traffic for unusual behavior or developing problems. The dental AI systems handling tasks like checking prior medical records or giving a second opinion on x-rays or helping with claims filing and record-keeping lets the dentist and dental technicians focus on delivering maximum value to patients.

Pearl is part of the Dental AI Council. The Dental AI Council aims to accelerate the adoption of AI in dentistry by exploring and validating its possible applications in dentistry, from early diagnosis to claims processing. The founding members of DAIC come from all parts of the dental industry, from practitioners, DSOs, manufacturers, practice management software providers, insurance carriers, laboratories and universities.

AI Compared to Dentist Diagosis

Pearl published a study comparing the diagnostic performance of three experienced human dentists to the performance of an artificial intelligence (AI) diagnostic system. They compared human and AI analyses of a set of 8,767 bitewing and periapical radiographs found the AI system to be more consistent and accurate at predicting the presence of tooth decay.

The study revealed inconsistency in human dentist diagnosis. While they showed moderate alignment – 79% unanimous agreement – when it came to the absence of decay, the three human dentists unanimously agreed about the presence of decay in only 370 X-rays––just 4.2% of the total. In nearly one in five instances, even when two dentists identified decay in an X-ray, the third dentist did not.

The superior accuracy of the AI system’s diagnostic conclusions remained consistent both when any single dentist’s diagnosis was held as ‘ground truth’, and when ground truth came from two dentists. The AI also tended to agree more with the human dentists than the human dentists did with each other, a product of the AI’s greater sensitivity to potential decay in X-rays.

The AI was trained on millions of X-rays which were evaluated by hundreds of human dentists. The AI is better at identifying tooth decay.

Dental Industry

The American Dental Association reports that there are over 200,000 dentists in the USA.

The global dental market size is about $30 billion per year and is growing at 5-7% per year.

7 thoughts on “Interview with the CEO of Pearl Which is Applying AI to Dentistry”

  1. Actually, I think you may have misunderstood the point of looking for potential fraud with AI. The insurance companies have to do it anyway, they're heavily regulated and audited, as many people rely on government insurance, and frankly they lose money on fraud.
    What the AI does is reduce the number of skilled people reviewing cases for potential fraud, freeing them up to perform other services for individuals. These aren't entry level positions, these are people who can read x-rays and understand diagnosis codes. Freeing up skilled workers tends to increase productivity and open up new opportunities in industries. Always a medium to long term good.

  2. While others, more conscientious people, can evaluate the overall meaning of what is brought here without becoming defensive.

  3. About 50% of dentists are still using amalgam fillings, which are pure mercury and dangerous

    And that's when I stopped reading.

Comments are closed.