Gingivitis is more common than expected, affecting roughly 46% of Americans over 30 years old (1). This is substantially higher than other common health problems such as: tension headaches (37%), cavities (25%), migraines (21%), and hearing loss (19%) (2). Gingivitis can lead to gum disease (also called “periodontitis”), which is found in 9% of the American population (1); The situation is even more dire elsewhere in the world (2). Gingivitis has recently been implicated as a cause for Alzheimer’s, as the gingivitis bacteria enters the brain (3).

In 2016, US dental costs totaled approximately $124 billion (4). Despite high costs, 82.2 million (~25%) Americans do not have dental insurance (5). This population is more at risk of painful and financially burdensome dental treatment. In addition, dentists are notoriously unpleasant and even individuals with insurance often go long stretches without any sort of care.

The goal of this project is to provide early-detection dental health assessments for individuals who are unable or unwilling to visit a dentist on a regular basis.

What it does

Perio-gone-titis is a rapid, low-cost hardware and deep learning platform to assess dental health. Our algorithm is trained on web-scraped dental and medical images and based on priors derived from an epidemiological model of American dental health using publicly-available data. To provide feedback to the user, we built a real-time video device with advanced signal processing of the user's mouth to identify gum-line inflammation invisible to the human eye but indicative of early gingivitis. We prototyped hardware at a cost of $2, greatly expanding accessibility to this valuable service.

Accomplishments that we're proud of

We synthesized a model combining methods from computer vision, signal processing, electrical engineering, and epidemiology to create highly accurate predictions.

Our project addresses an under-recognized public health issue by allowing users to assess more accurately their dental health without the time, expense, or discomfort associated with dental visits.

We prototyped a device at a cost of $2 that can detect gingivitis and related diseases with more than 80% accuracy.

What's next for Perio-gone-titis

We want to extend this platform to cavities, oral thrush, and canker sores. To do this, we plan to update our computer vision model to use propensity score matching of down-sampled images to enable future detection of numerous rare diseases of the mouth. Furthermore, we will add support for other organs, including ears and skin.

Eventually, we want to use this tool to improve public health worldwide. We will link users to dental hygiene educational material and on-call hygienists after they view their results. We also would like to mass mail devices to individuals without dental insurance as part of an awareness and triage campaign on the serious consequences of dental oversight.

Works Cited

  1. Eke PI, Dye BA, Wei L, Slade GD, Thornton‐Evans GO, Borgnakke WS, Taylor GW, Page RC, Beck JD, Genco RJ. Update on prevalence of periodontitis in adults in the United States: NHANES 2009 to 2012. Journal of periodontology. 2015 May;86(5):611-22.

  2. Vos T, Abajobir AA, Abate KH, Abbafati C, Abbas KM, Abd-Allah F, Abdulkader RS, Abdulle AM, Abebo TA, Abera SF, Aboyans V. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. The Lancet. 2017 Sep 16;390(10100):1211-59.

  3. Poole S, Singhrao SK, Kesavalu L, Curtis MA, Crean S. Determining the presence of periodontopathic virulence factors in short-term postmortem Alzheimer's disease brain tissue. Journal of Alzheimer's Disease. 2013 Jan 1;36(4):665-77.

  4. American Dental Association Health Policy Institute. “U.S. Dental Expenditures 2017 Update”. Available:

  5. US Department of Health and Human Services. Oral Health in America: A Report of the Surgeon General. Rockville, MD: US Department of Health and Human Services, National Institute of Dental and Craniofacial Research, National Institutes of Health; 2000.

Built With

Share this project: