UB dental school professor discusses how AI can advance periodontic treatment

I was always passionate about technology of all kinds. I never took a course in using AI, but during my residency, I would sit in front of the computer and figure out how to do it. I saw how it would make periodontic surgery faster and more precise. I started to love using it, and I wanted to teach others how to use it, too.

AI can be used in three different stages of care: diagnosis, treatment and maintenance. My work is mainly focused on diagnosis and treatment, such as the procedure on lengthening crowns. With AI, we precisely know where to cut the gum and where to remove the bone behind it. When you go freehand, it’s hard to get the measurements right. With AI, you do all the thinking process before the procedure and at the time of the procedure you just need to follow the guide.

For the crown-lengthening procedure, for example, we use AI to determine how the smile will look after the procedure. Once the patient signs off, we create a guide for the exact position where we need to cut. We print it and put it in the patient’s mouth. In another example, when doing an implant with a patient, a guide can show us the correct position and angulation of the implant. It’s particular helpful in cases with multiple implants.

It can highlight the treatments that the patient most likely needs by the X-rays, periodontal chart and the information that’s in the patient’s record. It can also tell you if the patient is missing a treatment or is due for a cleaning.

AI can also help diagnose a patient quicker and more accurately. For example, when an X-ray indicates a problem, there are a lot of variables. One clinician can look at the image and say, “There’s 50% bone loss,” while another clinician will say, “No, for me, it looks like 60%.”

I think AI is going to help us create a standard that will be the same for everybody.

My main goal is to teach students to use critical thinking while using AI tools. It is a technology in development and can make mistakes. Different kind of procedures have different kind of mistakes; we have to explore them so that we teach the machine how to do the procedures properly in the future.

The course is designed to supplement the clinical experiences of students in postgraduate programs in 3D surgical digital planning, microscope surgery and AI as applied to periodontics.

With the funding, I brought in experts from other schools who taught different aspects of AI. Four came in person and two presented the lectures virtually. They also worked with me on how to apply AI techniques on a larger scale and how to handle HIPAA (Health Insurance Portability and Accountability Act of 1996) regulations.

I was very happy with it. I initially planned for 11 periodontic residents, but in the end, about 25 residents from other dental programs, as well as periodontics faculty, came in to observe the class. I did a survey with students at the end, and they really liked the lectures and the guides because it allowed them to be up to date on new technologies. I can tell that the lessons worked because I see students using AI now. They’re digitally planning the surgical periodontal cases of real patients, using it for crown-lengthening procedures, advanced bone graft cases — like a sinus lift — and implant placement.

We plan to offer the course every other spring. The activities and access to new software and intraoral scanners are expensive. So, we will look for solutions and support to have it available to the students more often. The students indicated in the survey that they would like to learn more procedures using AI software. I also would like to introduce the AI-based software to dental students by adding this component to the periodontal course they already take and focus more on the diagnosis of the patient and risk assessment.