The Esthetic Perception of Morphological Severity in Scaphocephalic Patients is Correlated With Specific Head Geometrical Features.

OBJECTIVE: To investigate the relationship between perception of craniofacial deformity, geometric head features, and 3D head shape analyzed by statistical shape modeling (SSM). PATIENTS: A total of 18 unoperated patients with scaphocephaly (age = 5.2 ± 1.1m)-6 were followed-up after spring-assisted cranioplasty (SAC) (age = 9.6 ± 1.5m)-and 6 controls (age = 6.7 ± 2.5m). MAIN OUTCOME MEASURES: 3D head shapes were retrieved from 3D scans or computed tomography (CTs). Various geometrical features were measured: anterior and posterior prominence, take-off angle, average anterior and posterior lateral and horizontal curvatures, cranial index (CI) (cranial width over length), and turricephaly index (TI) (cranial height over length). SSM and principal component analysis (PCA) described shape variability. All models were 3D printed; the perception of deformity was blindly scored by 9 surgeons and 1 radiologist in terms of frontal bossing (FB), occipital bulleting (OB), biparietal narrowing (BN), low posterior vertex (LPV), and overall head shape (OHS). RESULTS: A moderate correlation was found between FB and anterior prominence (r = 0.56, P < .01) and take-off angle (r = - 0.57, P < .01). OB correlated with average posterior lateral curvature (r = 0.43, P < 0.01) similarly to BPN (r = 0.55, P < .01) and LPV (r = 0.43, P < .01). OHS showed strong correlation with CI (r = - 0.68, P < .01) and TI (r = 0.63, P< .01). SSM Mode 1 correlated with OHS (r = 0.66, p < .01) while Mode 3 correlated with FB (r = - 0.58, P < .01). CONCLUSIONS: Esthetic cranial appearance in craniofacial patients is correlated to specific geometric parameters and could be estimated using automated methods such as SSM.

Three-dimensional soft tissue prediction in orthognathic surgery: a clinical comparison of Dolphin, ProPlan CMF, and probabilistic finite element modelling.

Three-dimensional surgical planning is used widely in orthognathic surgery. Although numerous computer programs exist, the accuracy of soft tissue prediction remains uncertain. The purpose of this study was to compare the prediction accuracy of Dolphin, ProPlan CMF, and a probabilistic finite element method (PFEM). Seven patients (mean age 18years; five female) who had undergone Le Fort I osteotomy with preoperative and 1-year postoperative cone beam computed tomography (CBCT) were included. The three programs were used for soft tissue prediction using planned and postoperative maxillary position, and these were compared to postoperative CBCT. Accurate predictions were obtained with each program, indicated by root mean square distances: RMSDolphin=1.8±0.8mm, RMSProPlan=1.2±0.4mm, and RMSPFEM=1.3±0.4mm. Dolphin utilizes a landmark-based algorithm allowing for patient-specific bone-to-soft tissue ratios, which works well for cephalometric radiographs but has limited three-dimensional accuracy, whilst ProPlan and PFEM provide better three-dimensional predictions with continuous displacements. Patient or population-specific material properties can be defined in PFEM, while no soft tissue parameters are adjustable in ProPlan. Important clinical considerations are the topological differences between predictions due to the three algorithms, the non-negligible influence of the mismatch between planned and postoperative maxillary position, and the learning curve associated with sophisticated programs like PFEM.