A late arrival to healthcare’s AI revolution, dentistry is now leading it

Author: Dr. Kyle Stanley, Chief Clinical Office, Pearl

If you look back at the short history of artificial intelligence in health care, you’ll see that it’s blazed its brightest trail in fields where it supports life-saving treatment: oncology, cardiology, pulmonology, and the foundation of it all—radiology.

The clinical adoption of AI by radiologists has been swift compared to dentistry. According to the American College of Radiology, about 30% of radiologists now work with AI, and its use in the field is advancing. Brigham Mass General hospital has developed algorithms that can detect abdominal aortic aneurysms [1]. Harvard researchers have developed AI that detects disease on chest x-rays using natural language descriptions in clinical reports [2] and can do it as accurately as human experts [3].

“Over the next 10-15 years, we'll see more models become widely available and adopted, with the average radiologists practicing with 20-40 algorithms each, depending on their subspecialties,” predicts Dr. Keith Dreyer, the chief data science officer and vice chairman of radiology at the Mass General Brigham health system.

Dentists may be behind the curve on AI adoption, but they’re swiftly catching up.

Of course, by then, dentistry will have done the work of bringing AI diagnostics into the mainstream. Dentists may be behind the curve on adoption, but they’re swiftly catching up and bringing it to the frontline of patient care. In time, it’s dentistry that will make AI ubiquitous in healthcare. Here’s why.

AI encounters of the everyday kind

AI that detects a malignant brain tumor makes headlines. AI that detects on tooth 17, not so much. But according to the CDC, 90% of Americans aged 20 and up have had at least one cavity in their lifetime [4]. (25% have one right now. [5]) Thankfully, less than 1% will develop a brain tumor [6].

What that boils down to is that very few people will encounter AI within the confines of traditional medicine. Medical doctors don’t refer their patients for x-rays or scans until there is cause for concern. AI is called in to support life-or-death diagnoses.

Dental offices, on the other hand, perform x-rays every day, and patients get them taken at annual check-ups. In other words, well-care in dentistry starts with imagery. Since 65% of American adults (about 136 million people) visit the dentist every year [7], according to the CDC, they have a much higher chance of encountering AI-based diagnostic software while seated in the dentist’s chair.

AI-backed software available to dentists now can help them identify and diagnose pathologies that appear on x-rays. But they are also powerful tools for building patient trust. While patients get dental x-rays taken every year, they aren’t experts at reading them. According to a recent survey of dental patients, 65% said they don’t completely understand what their dentist is pointing out to them on those x-rays [8]. AI bounding boxes provide a focal point for patients struggling to see what their dentist is pointing to.

One truly outstanding feature of Pearl's Second Opinion® radiologic AI software is the tooth segmentation tool. It automatically distinguishes tooth parts and supporting structures, allowing dentists to effortlessly communicate their findings to patients.

This may be one reason why dentists are so quickly adopting the technology. While we’ve seen exponential growth at Pearl, there’s outside evidence that AI is making fast inroads. A 2022 study of how often artificial intelligence appears in dental literature showed a 21.6% increase between 2011 and 2021 and an even starker uptick of 34.5% over the last five years [9]. And of the publications selected from 2021, 26.36% of those references were to radiology, the day-to-day exposure most likely to popularize AI among patients [10].

Entrepreneurial adoption

Of course, AI can do much more for dentists and dental offices than just help read x-rays. They can help with treatment planning, margin calculations, and back-office operations. Unlike doctors who work in hospital systems, 73% of dentists own their practice, according to the American Dental Association [11]. The significance of that is two-fold.

First, they have a financial interest in maximizing treatment and production. Empty schedules mean patients aren’t getting the treatment they need—and that practices aren’t using their hours efficiently. When artificial intelligence gives the front office an edge in identifying urgent and high-need cases that fill out the weekly schedule, practice owners are likely to consider it.

Secondly, purchasing AI tools within medical systems is subject to lengthy panel reviews, RPF processes, and vendor evaluations. The slow march of bureaucracy impedes the adoption of everyday tools by physicians. Meanwhile, dental practice owners need only consult themselves or their partners when deciding if they should try AI-powered software.

More and more, they’re choosing to do so. That choice will make dentistry the field in which AI is most ubiquitous.

If you are located in Australia or New Zealand and want to learn more about Pearl's Second Opinion®, contact our team at Gamma Tech.

---

[1] Mass general Brigham and the future of AI in Radiology. Healthcare IT News. (2021, May 10). Retrieved November 7, 2022, from https://www.healthcareitnews.com/news/mass-general-brigham-and-future-ai-radiology

[2] Pesheva, E. (2022, September 15). No labels? No problem! Harvard Medical School. Retrieved November 7, 2022, from https://hms.harvard.edu/news/no-labels-no-problem

[3] Tiu, E., Talius, E., Patel, P., Langlotz, C. P., Ng, A. Y., &; Rajpurkar, P. (2022, September 15). Expert-level detection of pathologies from unannotated chest X-ray images via self-supervised learning. Nature News. Retrieved November 7, 2022, from https://www.nature.com/articles/s41551-022-00936-9

[4] Oral Health Surveillance Report: Trends in Dental Caries and Sealants, Tooth Retention, and Edentulism, United States, 1999–2004 to 2011–2016. Atlanta, GA: Centers for Disease Control and Prevention, US Dept of Health and Human Services; 2019, from https://www.cdc.gov/oralhealth/publications/OHSR-2019-index.html

[5] Ibid.

[6] Key statistics for brain and spinal cord tumors. American Cancer Society. Retrieved November 7, 2022, from https://www.cancer.org/cancer/brain-spinal-cord-tumors-adults/about/key-statistics.html

[7] National Health Interview Survey. National Center for Health Statistics. (2019.)

[8] Dental Patient Trust & Technology Survey: Technology’s role in building trust among dental patients. Pearl. (2022.).. Retrieved November 7, 2022, from https://www.hellopearl.com/guides/patient-trust.

[9] Thurzo A, Urbanová W, Novák B, Czako L, Siebert T, Stano P, Mareková S, Fountoulaki G, Kosnáčová H, Varga I. Where Is the Artificial Intelligence Applied in Dentistry? Systematic Review and Literature Analysis. Healthcare (Basel). 2022 Jul 8;10(7):1269. doi: 10.3390/healthcare10071269. PMID: 35885796; PMCID: PMC9320442.

[10] Ibid.

[11]Practice Ownership Among Dentists Continues to Decline [Infographic]. ADA Health Policy Institute. (2021.), from https://www.ada.org/-/media/project/ada-organization/ada/ada-org/files/resources/research/hpi/hpigraphic_practice_ownership_among_dentists_decline.pdf