The influence of artificial intelligence in orthodontics.

Fraser Macdonald, Dental Tribune International

Tue. 8. April 2025

save

LEIPZIG, Germany: Borne by the currents of economic efficiency, time-saving and convenience, artificial intelligence (AI) continues its steady yet rigorous transformation of the dental world. Nowhere has this transformation been expressed more potently than in the realm of diagnostics, where a range of AI-powered platforms are capable of accurately and instantaneously assessing a wide range of visual information for specific dental conditions. It is no surprise that orthodontics is being powerfully influenced by the functionalities of AI. In this article, I outline some of the key applications in this sphere and ponder the fundamental and interrelated questions of accuracy and human redundancy. Fortunately, despite popular fears that our lives will be usurped by robots, the role of the clinician continues to be essential.

AI's potential impact on orthodontics is most notable in diagnostics, where it has shown good capability in automating labour-intensive processes and enhancing precision. According to a recent scoping review published in the Journal of Dentistry, various studies have demonstrated the application of AI in cephalometric analysis, the diagnosis of occlusal traits, maturation determination and upper airway assessment. For instance, some AI models have achieved accuracies exceeding 90% in tasks like cervical vertebral maturation staging. Similarly, AI has shown remarkable performance in diagnosing malocclusions using 2D and 3D imaging, achieving accuracies of up to 99% in specific cases. These advancements significantly reduce the time required for manual evaluations and mitigate the variability inherent in human assessments.

The advancements made by AI in the realm of treatment planning have been no less significant. Employing the same capacity to instantaneously and precisely assess visual information, AI facilitates the subsequent evaluative step in the clinical workflow. In orthodontic treatment planning, AI has proved invaluable in streamlining complex decision-making processes. AI models have been employed to predict the need for tooth extractions, plan orthognathic surgeries and assist in general orthodontic workflows. Studies have reported accuracies of over 93% in determining whether extractions are necessary, as well as shown precise segmentation of anatomical structures using 3D imaging. Additionally, AI models have been developed to simulate treatment outcomes, predict patients’ experiences of aligner treatment regarding pain, anxiety and quality of life, and prioritise orthodontic issues for resolution.

Despite their obvious and important benefits, AI models still require human supervision to ensure reliability.

A powerful impetus for the use of AI in dentistry has been its ability to undertake analytical tasks in a fraction of the time and, ideally, with heightened accuracy and efficiency. In orthodontics, for example, tasks such as identifying cephalometric landmarks and conducting cervical vertebral maturation staging are time-intensive and prone to human error. AI models can automate these processes within seconds, offering results that are often comparable to those of expert clinicians. However, overall, the scientific consensus remains that the accuracy of these models has not yet consistently exceeded assessments by expert clinicians, a perspective supported by both systematic reviews and professional opinion. Despite their obvious and important benefit

Dr Soumya Narayani Thirumoorthy. (Image: Dr Soumya Narayani Thirumoorthy)

s, AI models still require human supervision to ensure reliability, as their diagnostic and analytical capabilities are still under development. For instance, certain models excel in simple cases like fully dentate patients but struggle with more complex scenarios, such as partially edentulous patients or those with craniofacial deformities. Another limitation is the current models’ reliance on 2D radiographic data, which may not accurately represent 3D anatomical structures. Although there is growing interest in applying AI to 3D imaging, studies in this area remain limited.

The importance of retaining human oversight in the increasingly digitised areas of diagnostics and treatment planning is the topic of a recent article in Evidence-Based Dentistry, carrying the subtitle “Are we ready to let Mr. Data run our Enterprise?”. Co-author of the article Dr Soumya Narayani Thirumoorthy, a board-certified orthodontist at SmileLife Orthodontics in Texas in the US, shared her perspective on the matter with Dental Tribune International: “AI significantly saves time in aligner treatment planning by providing simulated results that serve as an enhanced starting point for the orthodontist to modify and finalise. However, the final decision rests with the orthodontist, as each patient has unique needs that require a personalised approach. Orthodontics is not a one-size-fits-all field, and the expertise of the human orthodontist is essential in achieving an outcome that best meets the needs of each individual patient. Therefore, AI should be seen as an assistant that enhances efficiency, but the orthodontist’s experience and ability to adjust treatment plans based on patient input remains crucial.”

Besides the efficacy of AI in orthodontics, Dr Thirumoorthy spoke about the future directions of AI in the field. “There are still many under-explored areas in our field that could benefit from AI. A few possibilities include predicting orthodontic relapse in patients and forecasting the development of white spot lesions. I envision a future where AI acts as the eyes and ears of the orthodontist, assisting with time-consuming tasks like data analysis, allowing clinicians to use their time more efficiently and focus on direct patient care,” she said.

The already powerful functionality of AI in orthodontics appears set to continually improve as its diagnostic and treatment planning algorithms are iteratively strengthened. The role of the clinician being on firm ground, it seems that the future holds the promise of a fruitful synergy between human and technology.