Abstract
Craniosynostosis is a congenital defect which can result in abnormal cranial morphology. Three dimensional (3D) stereophotogrammetry is potentially an ideal technique for the evaluation of cranial morphology and diagnosis of craniosynostosis because it is fast and harmless. This study presents a new method for objective characterization of the morphological abnormalities of scaphocephaly and trigonocephaly patients using 3D photographs of patients and healthy controls. Sixty 3D photographs of healthy controls in the age range of 3–6 months were superimposed and scaled. Principal component analysis (PCA) was applied to find the mean cranial shape and the cranial shape variation in this normal population. 3D photographs of 20 scaphocephaly and 20 trigonocephaly patients were analysed by this PCA model to test whether cranial deformities of scaphocephaly and trigonocephaly patients could be objectively identified. PCA was used to find the mean cranial shape and the cranial shape variation in the normal population. The PCA model was able to significantly distinguish scaphocephaly and trigonocephaly patients from the normal population. 3D stereophotogrammetry in combination with the presented method can be used to objectively identify and classify the cranial shape of healthy newborns, scaphocephaly and trigonocephaly patients.
Craniosynostosis is a congenital defect that causes one or more sutures of the newborn’s skull to ossify prematurely. As a result, normal skull growth is impeded in the direction perpendicular to the affected sutures, resulting in abnormal cranial morphology. In addition to abnormal skull growth, craniosynostosis can result in complications such as intracranial hypertension, visual impairment, and limitation of brain growth .
To prevent or resolve these complications, surgical treatment is required. Today, a variety of surgical treatment options are available. The most commonly used types of surgery for patients with a single-suture craniosynostosis include cranial vault remodelling, spring-mediated cranioplasty and endoscopically assisted strip craniectomy followed by remoulding helmet therapy .
Because there is still a lack of a proper validated method to evaluate the cranial vault of craniosynostosis patients and to establish an objective quantification of the severity of the abnormality, there is no consensus on which of the treatment options provides the best long-term functional and aesthetic outcome. Therefore, the follow-up of craniosynostosis treatment and finally the determination of the optimal surgical procedure remains a rather subjective matter . Three-dimensional (3D) stereophotogrammetry is a fast and patient-friendly method to evaluate the complete 3D morphology of the cranial shape. Because no potentially harmful ionizing radiation is required, 3D stereophotogrammetry is an ideal technique to acquire a 3D image of the cranial shape for diagnosis and during follow-up.
The aim of this study was to present a new method for the objective evaluation of the 3D morphology of the cranial shape. 3D photographs of a healthy control group were used to establish the normal cranial shape and its variations. Scaphocephaly and trigonocephaly patients were compared with the controls and principal component analysis (PCA) was used to test if there was a significant difference between the groups.
Patients and Methods
Subjects
A total of 100 3D stereophotographs were included in this prospective study. This dataset included 20 3D photographs of patients with scaphocephaly and 20 3D photographs of patients with trigonocephaly. The included scaphocephaly and trigonocephaly patients suffering from isolated, non-syndromic premature closure of respectively the sagittal or the metopic suture. Diagnosis was confirmed using a computed tomography (CT) scan. The pre-operative 3D photographs of the scaphocephaly patients were taken at an average age of 4.1 months (range 2.9–5.1 months) and the 3D photographs of the trigonocephaly patients at an average age of 4.0 months (range 3.0–5.1 months).
A total of 60 3D photographs of healthy controls were included in the age range of 3–6 months. All control subjects were without known associated anomalies.
Data acquisition
Three-dimensional stereophotogrammetry (3dMDCranial 3DMD, Atlanta, USA) with a five-pod configuration was used for image acquisition. The 3D photographs of all subjects were acquired by a trained photographer. In case the reflection of hair caused artefacts on the 3D photograph, the subjects were photographed with a tight nylon skull cap. All 3D photographs of the patients were acquired one day prior to surgery. Severe quality inconsistency in the acquired 3D photographs was an exclusion criterion.
Superimposition of 3D photographs
For adequate analysis, every 3D photograph needed to be positioned in the same coordinate system and therefore rotations and translations needed to be eliminated. To achieve this, all 3D photographs were imported into the 3DMDPatient software (3dMD, Atlanta, GA, USA). Eight anatomical landmarks were manually identified on the 3D photographs ( Table 1 ). The 3D photographs and landmarks were subsequently exported to the technical computing software Matlab (MATLAB v2012b, The Mathworks Inc., Natick, MA, USA). In Matlab, a horizontal plane defined by the exocanthion, endocantion, and pretragion points and a vertical plane, perpendicular to the horizontal plane and positioned through the subnasal point were generated. Based on the method described by Brons et al., the planes were used to eliminate the rotations ( Fig. 1 ). All 3D photographs were translated to the origin of the reference frame using the computed cranial focal point (CCFP) presented by de Jong et al. . This resulted in a dataset of 3D stereophotographs in the same coordinate system without rotational and translational differences.
