Evaluation of Artificial Intelligence in Planning the Treatment for Cases With Missing Upper Lateral Incisors (A Diagnostic Accuracy Pilot Study)

Management of congenitally missing maxillary lateral incisors is a frequent clinical challenge in orthodontics, often sparking debate among practitioners regarding the optimal treatment approach. The two primary treatment modalities are orthodontic space closure, in which adjacent teeth (usually canines) are moved into the lateral incisor position, and space opening followed by prosthetic replacement, typically with dental implants or bridges. Each option has distinct advantages and limitations in terms of esthetics, occlusion, periodontal health, and long-term stability, and the decision is influenced by multiple clinical, biological, esthetic, and patient-specific factors.

Artificial intelligence (AI), particularly through machine learning algorithms, has demonstrated high accuracy in other controversial orthodontic decision-making scenarios, such as extraction vs. non-extraction treatment and surgical vs. camouflage approaches. However, its application to lateral incisor agenesis treatment planning has not been thoroughly investigated.

This pilot diagnostic accuracy study will evaluate the performance of an AI-based decision support system in recommending treatment plans for cases with missing maxillary lateral incisors. A dataset of anticipated 100 cases will be compiled, consisting of pre-treatment records including intraoral and extraoral photographs, panoramic radiographs, and cephalometric analyses. two experienced orthodontists will independently review unfinished cases and make a treatment decision-space closure, space opening with prosthetic replacement, or undecided-using a standardized cutoff-points checklist. For finished cases, both pre- and post-treatment records will be analyzed.

A consensus decision will be established when at least two orthodontists agree; if disagreement persists, a third orthodontist will finalize the decision. AI predictions will be compared with orthodontists' consensus decisions to assess diagnostic accuracy, sensitivity, specificity, and agreement rates. This study aims to explore the feasibility of integrating AI tools into complex orthodontic decision-making and to establish a foundation for larger-scale clinical trials.