OBJECTIVE: To verify the accuracy of a digital workflow for occlusal relationship reduction in patients with maxillofacial fractures via cone beam computed tomograhy/ computerized tomography (CBCT/CT). METHODS: We collected CBCT/CT data and oral scan data from both general patients and fracture patients and established digital models. The differences between the oral scan model and the CBCT/CT digital model were compared. 3D-printed dental models were made for general patients, and comparisons were made matching the dental model with the original occlusion. Differences among the postoperative CT digital models of fracture patients, the virtual reconstruction models, and the model surgery reconstructed models were compared. RESULTS: A total of 20 general patients and 27 fracture patients participated in this study. The average error for the segmented CBCT/CT modelling group was 122.90±26.94 μm, whereas the average error for the nonsegmented CBCT/CT modelling group was 84.50±9.21 μm, with a significant difference between the two groups (t = 6.364, P <
0.01). The error in the occlusal relationship between the 3D-printed models and the patient's actual occlusion was 18.35±2.20 μm. The average error for the virtual reconstruction group was 438.89±155.55 μm, whereas it was 857.09±116.55 μm for the model surgery reconstruction group, with a significant difference between the two groups (t = 11.180, P <
0.01). CONCLUSION: CBCT/CT-based digital dental models can be created with high accuracy. Using 3D-printed models and intercuspal occlusion guidance, the original occlusal relationship can be restored. Compared with model surgery, virtual fracture reconstruction offers superior accuracy. CLINICAL SIGNIFICANCE: This study demonstrates the feasibility of accurate digital occlusal restoration in fracture patients, providing a new approach for maxillofacial fracture management.