United Reference Method for three-dimensional treatment evaluation.

BACKGROUND: Reproducible and descriptive Three-dimensional treatment evaluation can enhance future treatment based on realistic results. So, the purpose of this study was to describe a new method for three-dimensional treatment evaluation showing how to use fully-automated craniofacial superimposition of CBCT records combined with reference system to obtain descriptive and comparable results. This new method was named United Reference Method (URM).

METHODS: URM is a combination of automated 3D superimposition on anterior cranial base surface anatomy and measurements based on reference system. It was developed to show how to use fully-automated superimposition to obtain descriptive numerical comparable values. The method is based on: one main reference system for both superimposed CBCT records, semi-automation to increase accuracy, all measurements are projections and auxiliary references to aid in landmarks identification and measurements. The method steps can be described following a four-step approach: (1) Superimposition performed through a fully automated, voxel-wise, rigid registration considering only cranial base as a stable structure; (2) Identification of reference landmarks once on the superimposed records for corrected Frankfort Horizontal plane (C-FH) construction and a new semi-automated constructed Sella point to correct Orbital asymmetry; (3) Head orientation of superimposed CBCT images based on the C-FH; (4) Identification of landmarks affected by treatment with the aid of auxiliary reference planes. Evaluation of linear or angular changes derived by projection of same pre- and post-treatment landmarks on the C-FH. Pre- and post-expansion CBCT scans of 20 unilateral cleft lip and palate patients were used to calculate intra and inter-rater reliability. (X, Y and Z) coordinates, mean, standard deviation (SD) and Intra-class Correlation Coefficient (ICC) were calculated.

RESULTS: The proposed coordinates for C-FH construction showed ICC ≥ 0.998 and SD ranging from 0.064 to 0.242 mm. On the other hand, excluded coordinates due to expected natural craniofacial asymmetry had the lowest reliability ICC ≥0.742 and SD dramatically increased up to 1.112 mm.

CONCLUSION: URM showed adequate reliability so it can be used to produce three-dimensional descriptive data of craniofacial structural changes.