Interview with Prof Falk Schwendicke – "The future of dentistry lies in predictive, personalized, and data-driven care"

Prof. Falk Schwendicke is a renowned expert in digital health and dental diagnostics, known for his pioneering work in applying artificial intelligence (AI) to dental care, image analysis, and treatment planning. (Photo: Prof Falk Schwendicke)

Prof. Schwendicke, your lecture at the 37th Dental ConfEx in Dubai is titled "Artificial Intelligence in Dentistry: Are we there yet?" What can delegates expect to learn from your presentation, and how do you see AI reshaping the future of dental diagnostics and treatment?
In my presentation, I will first explain how AI technology works, as there’s a lot of talking around it, but many people don’t really understand its mechanics. This is important because the same technology behind AI in dentistry powers things like your iPhone, passport automation, and even self-driving cars. After outlining the fundamentals, I’ll dive into the current applications in dental image analysis, such as radiographs, scans, and photos, particularly in practices focused on aesthetics, smile design, and complex rehabilitation, where planning on 3D and 2D images is crucial.

Additionally, I will cover how AI applies to speech, text, and numerical data. Large foundational models like ChatGPT or predictive models can help in patient communication, administration, and dental logistics. Attendees will gain an understanding of AI’s potential impacts across a broad spectrum of dental practice processes.

You’ve had a remarkable career in dental diagnostics, preventive care, and health economics. How did AI become a central focus of your work, and what initially drew you to this area of research?
It's quite a funny story. I was involved in a health services research project, developing an atlas for Germany's dental density and service provisions for the government. During that time, I had a statistically inclined staff member, and when machine learning first emerged in fields like dermatology and ophthalmology, I came across one of those papers and thought, "Could we do this with our radiographs?" I showed it to my colleague, and he said, "Yes, we can."

That moment led us to join an accelerator program at Berlin University, where I was a professor at that time. Eventually, we founded a company, and today I find myself chairing numerous initiatives at the World Health Organization, ISO, FDI, and others. It was all a bit of an accident, but we happened to be the first to apply AI in dentistry in academia.

One of the key points in your lecture is discussing the limitations and risks of AI in dentistry. What are some of the current barriers preventing widespread AI integration into dental practice?
There are two main challenges. First, the AI systems we have today are relatively young with many limitations—most are just a year or two old—so there are concerns about their generalizability, robustness, and the transparency of the data they use. These are areas we are actively working on, for instance, with the WHO ITU Focus Group AI for Health, to improve the reliability and accessibility of AI systems.

The second challenge lies within dental practices. Many dentists lack the necessary digital literacy to implement these tools. This is where events like the 37th Int’l Dental ConfEx become vital, as they help bridge this gap. Additionally, in dentistry, we face significant issues with data interoperability. Our datasets aren’t standardized and often exist in silos, making it difficult to integrate AI systems with existing practice management software. These are the two major hurdles slowing progress—not stopping it, just slowing it.

Your involvement with dentalXrai GmbH has been pivotal in advancing AI-driven image analysis in dentistry. Could you share the story behind this spin-off and its impact on the field?
As I mentioned earlier, this all began with a research collaboration between me and my colleague, Joachim, back in 2017-18. We entered an accelerator program that was designed to help researchers spin off companies, and by 2020, we founded dentalXrai GmbH, though we knew very little about running a business at the time. It was an adventure, and we probably made a lot of mistakes along the way.

In 2022, we were fortunate that Align Technology acquired our company, and today, our software is used in thousands of practices across Europe. It has had a meaningful impact, bringing AI-driven image analysis to many dental professionals.

The concept of P4 dentistry—predictive, preventive, personalized, and participatory—is intriguing. How do you see AI playing a role in the broader adoption of this model in everyday dental care?
AI is one tool that will drive the adoption of P4 dentistry. While AI won’t be the central focus, it will enable more predictive and personalized care, provided we can access the right data. Our current datasets in dentistry are too small to support the large models we use in other fields, such as ChatGPT. To truly harness AI, we need larger, more diverse data pools. This means opening up the various data silos in dentistry and thinking about federated learning, where we can collaborate across organizations to build better models. 

Finally, looking forward, how do you envision AI transforming dentistry over the next decade? Are we on the brink of a major shift, or do you think the transition will be more gradual?
I think we’re already seeing some of the transformations. In image diagnostics, AI will soon become the standard. Within the next five to ten years, I believe most dentists in high-income countries will rely on AI to assess images. We’re also seeing scanners becoming central to patient documentation, where AI overlays help create a “digital twin” of the patient.

In addition to diagnostic tools, AI will help shift dentistry toward more predictive, individualized care—the P4 model. Another major trend is conversational AI. In the near future, we’ll see dental assistants, admin staff, and even dentists interacting with AI through natural speech. Tasks like booking appointments, rescheduling, charting, and even patient education will be automated. This will revolutionize how we run practices, making things more efficient and improving patient care.

Thank you.