How Artificial Intelligence Works
In Dentistry

This photo depicts a machine learning image. Because the AI algorithm can identify disease, the pink region highlights enamel and the green shading is a segmented identification that shows the clinician the extent of the decay. Some models will use a bounding box so the clinician can easily visualize decay. This AI algorithm can also help the clinician calculate decayed, missing and filled surfaces.

This panoramic image depicts machine learning identification of each tooth to help the clinician identify the teeth, anatomy and tooth number. With AI technology, the computer has been trained through neural network applications and millions of X-rays to aid the clinician with this kind of identification and color enhancements.

The first image in this sequence shows a standard X-ray, while the second shows the same radiograph with AI enhanced, color-coded information. The AI-enhanced image offers the clinician clear views of the enamel mask, tooth numbers, measurement from the cementoenamel junction to the crestal bone (measured to the 10th of a millimeter). Note that restorations can also be identified.

Similar to how certain tools can look at the world in infrared, UV, heat-mapping, AI algorithms have the ability to look at images in ways we as humans are unable to. These are only 4 examples of many possible AI models.

This photo depicts a sample dashboard. Not only does the AI algorithm identify images relevant for the clinician, but enhancements including bounding boxes and shading will help identify bone loss, decay or compressed margin. CEJ and crestal bone measurements also offer clinical decision support to aid diagnostic and treatment planning.

The first image shows the radiograph of a root canal, and the second AI-enhanced image identifies residual infection around the root.

To determine whether or not the root canal is well obturated, this AI-enhanced image shows the root canal fill in green and helps the clinician measure the working length (fill to the tip of the root).

This AI-enhanced radiograph identifies the crowns (dark blue), restorations (purple) and implants (far right). This image also shows tooth numbers and bone loss measurements. The AI software is flexible enough to let the clinician toggle on/off the color-enhanced functionality to aid in the diagnostic evaluation of a patient.

This image shows assessment of marginal seal around crowns, pre-cementation. Obtruation assessment and tracking of potential pathology over time is one AI benefit. We also see recurrent decay around crown margins and direct restorations. We can track incipient lesions over time following fluoride treatment and hygiene regimen. 

In this radiograph, the AI-enhancements clearly show decay (green) under the enamel (pink) and through the dentin. The algorithm can also identify decay that is has not yet broken through the enamel to help the clinician decide whether remineralization is a treatment option.