Home > Technology peripherals > AI > body text

How AI can help dentists solve hard-to-find dental problems

王林
Release: 2023-04-13 09:22:05
forward
909 people have browsed it

​Healthcare is in the midst of a digital revolution. Patients are already familiar with electronic record keeping and digital CT and MRI scans; some use computer vision and artificial intelligence (AI) to diagnose lung and other cancers.

How AI can help dentists solve hard-to-find dental problems

Although artificial intelligence is ready to bring new and more direct impressions to our lives, in daily life Dentistry is where most people probably first experience the amazing power of computers - their ability to view and interpret routine dental X-ray reports more accurately than humans can.

West Hollywood startup Pearl provides artificial intelligence to dental images to assist with diagnosis. It received FDA approval in March this year and is one of the first companies to receive approval for dental AI.

The approval paves the way for its use in clinics across the United States.

“This is truly a first in dentistry,” said Ophir Tanz, co-founder and CEO of Pearl. “But we have similar regulatory approvals in 50 countries around the world.”

The Importance of Artificial Intelligence in Dentistry

Dentistry can make good use of the capabilities of artificial intelligence. Everyone who goes to the dentist has an X-ray taken from time to time, so there are probably many more dental X-rays in existence than for any other type of disease. These radiographic images, annotated by human experts, are used to teach the AI ​​system what healthy and unhealthy teeth look like and how to recognize the difference.

Second Opinion is an AI detection platform created by Pearl, a dental startup founded in 2019 to use machine learning and AI to help dentists detect problems in healthy teeth. The startup raised $11 million in Series A funding in 2019 from Craft Ventures and Santa Monica-based Crosscut Ventures.

To develop Second Opinion, Pearl collected more than 100 million dental X-rays from dental offices and academic institutions. The AI ​​platform points out differences found in X-rays and also serves as a patient communication tool, allowing dentists to show different models of a patient’s teeth and pinpoint problem areas.

“I do think this will quickly become the foundation for the [dentistry] category, and so will actually be a model for other areas of medicine—how to infuse and deploy artificial intelligence at scale, in a truly demonstrable way. The ultimate benefit and potential to improve the standard of care," said Ophir Tanz, Pearl founder and CEO.

Pearl’s software platform, available as a cloud service, enables dentists to perform real-time screening of X-rays. Dentists can then review the AI ​​findings and share them with patients to facilitate informed discussions between dentists and patients about diagnosis and treatment plans.

Behind the scenes, NVIDIA GPU-powered convolutional neural networks developed by Pearl can spot not just cavities, but many other dental problems, such as cracked crowns and root abscesses that require root canals.

Pearl’s AI delivers dental results. The startup’s FDA filing shows that Pearl AI can spot 36% more pathologies and other dental problems on average than a typical dentist. "This is important because in dentistry, missing pathology is very common and routine," Tanz said.

The company's products include its Practice Intelligence, which enables dental offices to run artificial intelligence on patient data to uncover missed diagnostic and treatment opportunities. Pearl Protect can help screen for dental insurance fraud, waste and abuse, while Claims Review provides automated claims checking.

The Key Behind Artificial Intelligence

Pearl’s founders labeled over a million images to help train their proprietary CNN model for a range of conditions common in dental offices. The model is run on NVIDIA V100 Tensor Core GPUs in the cloud to identify the problem. Prior to this, they prototyped on local NVIDIA-powered workstations.

Inference is done on cloud-based GPUs, and Pearl’s system synchronizes with the dentist’s real-time and historical radiology data. “The dental vertical is still transitioning to the cloud, and now we’re bringing artificial intelligence to the cloud—we represent a wave of technology that will drive dentistry into the future,” Tanz said.

Getting FDA approval won't be easy, he said. It requires completion of an extensive clinical trial. Pearl submitted four studies, each involving thousands of X-rays and more than 80 expert dentists and radiologists.

“We are the only company in the world that can diagnose pathology and detect disease in a dental practice in an AI-driven way,” he said. "We are pushing for a more comprehensive diagnosis and it is a diagnostic aid for general practitioners."

Promote doctor-patient relationship

The dentist examines the , and feel free to point out any issues. Even for experienced dentists, understanding the grayscale images that form the basis of most treatment plans can be challenging, and for patients, understanding the blurred grades in X-rays that separate healthy tooth structure from unhealthy tooth structure is even more challenging. difficulty.

But with AI-assisted images, dentists can outline areas of concern with easy-to-understand bounding boxes. This ensures that their treatment plan has a sound basis while giving patients a clearer understanding of what exactly is happening with their X-rays.

"You can have highly visual discussions and paint a visual narrative for the patient so that they really start to understand what's going on in their mouth," says Dr. Tanz.

The above is the detailed content of How AI can help dentists solve hard-to-find dental problems. For more information, please follow other related articles on the PHP Chinese website!

Related labels:
source:51cto.com
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template
About us Disclaimer Sitemap
php.cn:Public welfare online PHP training,Help PHP learners grow quickly!