Abstract

Statement of the Problem: Accurate determination of maxillomandibular relationships is essential in prosthodontics but conventional methods are technique-sensitive and time-consuming. The accuracy and precision of AI-driven digital articulation remain underexplored.

Purpose and objective: This study aimed to evaluate the accuracy and precision of an AI-based software (Bite Finder) in locating maxillomandibular relationships from digital scans compared to traditional hand-mounted casts, VPS bite registration, and intraoral scan (IOS) methods.

Materials and Methods: Four groups were tested: (1) hand-articulated casts, (2) VPS bite-registered casts, (3) IOS scans with occlusal registration, and (4) AI-articulated digital models. Dental casts were fabricated and scanned, while IOS scans were acquired from participants. Inter-arch linear measurements between anatomical landmarks were compared to a reference dataset. Data normality was assessed with the Shapiro-Wilk test, followed by non-parametric two-way ANOVA and General Linear Model analyses.

Results: No significant differences were found between the Control and IOS groups across all landmarks (p > 0.05). The Control, VPS, and AI groups showed identical measurements, indicating high consistency. Statistical tests confirmed comparable accuracy and precision across all methods without systematic bias.

Conclusion: AI-based articulation software demonstrated accuracy and precision comparable to conventional and IOS-based methods in establishing maxillomandibular relationships. These findings support the clinical integration of AI-driven digital workflows as reliable alternatives, offering improved efficiency without compromising diagnostic quality.