Medical imaging techniques require various body positions. Gravity causes changes in the facial soft tissue and acts in different directions according to the position of the head during imaging. The aim of this study was to evaluate the effect of positional changes on the facial soft tissue. The faces of subjects were scanned in the standing, sitting, and supine body positions. Differences in the positions were compared using the root mean square (RMS), mean absolute deviation (MAD), and mean signed distance (MSD). The displacement of 15 midsagittal and 20 bilateral landmarks was evaluated. The RMS, MAD, and MSD values of the sitting–standing comparison were significantly lower than those of the sitting–supine and standing–supine comparisons. There were no significant differences between the sitting–supine and standing–supine comparisons. Sixteen out of 135 measurements (12%) of the midsagittal landmarks and 94 out of 180 (52%) measurements of the bilateral landmarks showed significant displacements among the body positions. These results demonstrate a significant change in the facial soft tissue caused by body position. Furthermore, these data show the different susceptibilities of the facial soft tissue landmarks to the effect of body position along the x , y , and z axes.
The assessment of the facial soft tissues is important for medical evaluations both before and after surgery. Because of the behavioural complexity of the facial soft tissues, three-dimensional (3D) evaluations are necessary in order to obtain more accurate results. The superimposition of facial masks is often essential to compare or analyze data. As well as clinical imaging techniques, novel 3D imaging technologies such as 3D digitization, 3D photography, laser scanning, and light scanning are becoming more popular among maxillofacial surgeons, plastic surgeons, and orthodontists because of their portability, ease of use, non-invasiveness, and high accuracy, as well as the availability of several complicated analysis options provided by the manufacturers.
During cephalometric analysis, the patient’s head orientation varies depending on the technique used, from lying down, to sitting, to standing. For instance, magnetic resonance imaging (MRI) and computed tomography (CT) scans are performed with the head in a horizontal position, while radiography, cone beam computed tomography (CBCT), 3D photography, and laser scanning are performed with the patient seated or standing. Each position has a different impact on the facial soft tissue due to gravity and causes a change in the facial expression. The use of such miscellaneous data sources can cause confusion when comparing the data sets from either the same or different studies.
The aim of this study was to evaluate the effect of the sitting, standing, and supine body positions on the facial soft tissue by analyzing the facial mask and the most commonly used landmarks in three dimensions.
Materials and methods
3D face scans of 70 volunteers were used in this study. The volunteers were 35 females and 35 males aged between and years (mean age . years). All of the volunteers were of Caucasian ethnic origin and had not undergone any previous orthodontic or surgical treatment.
Thirty-five landmarks were used. Fifteen of the landmarks were identified in the midsagittal plane ( n = 15) and were bilateral ( n = 20). Most of these landmarks were defined according to Farkas ( Fig. 1 ). The landmarks were marked with a pen on the skin of the volunteers by the same observer in order to compare exactly the same point in the different head positions. After this procedure, the subjects were scanned in the sitting, standing, and supine body positions with a 3D light scanner.
An Artec EVA 3D light scanner (2013; Artec Group, Luxembourg) was used for the face scanning. The working distance of the scanner is between 0.4 m and 1 m, the 3D accuracy is up to 0.05 mm, and the 3D resolution is up to 0.1 mm. During scanning, a computer and Artec EVA Studio software (version . . .15; Artec Group) was used to host the Artec EVA. Scanning was performed in the sitting, standing, and supine positions. The sitting and standing scans were performed in the natural head position. The natural head position was determined by the subjects own feeling of a natural head balance. The supine scans were performed after the head of the subject was oriented horizontally ( Fig. 2 ).
Soft tissue analyses
The scanned data from the three positions were imported into the same workspace of the Artec Studio software. The corresponding scans for each position were aligned manually using translation and rotation in all three directions, and then merged onto each other using the surface mapping algorithm. For alignment, the mid-endocanthion point (a point halfway between the inner corners of the eyes), which has been found to be the statistically most stable point of the face, was used as the common origin of the x , y , and z axes. The transverse, vertical, and sagittal anatomical axes of the head were defined as the x -, y -, and z -axis, respectively. The marked landmarks were used to bring the scans into the same position and give them the same orientation. For each landmark, the differences between the sitting and supine, standing and supine, and sitting and standing positions were calculated along the x , y , and z axes. In addition, the standard deviations (SD) of the landmarks were obtained for the 35 landmarks. The root mean square (RMS), the mean absolute deviation (MAD), and the mean signed distance (MSD) values were also calculated automatically by the software; these were then used to compare the superimposed scans. Figure 3 shows an example of the deviation colour maps derived from the superimposition of facial masks obtained from the scans.
The statistical analysis was performed using GraphPad Prism version .05 software (GraphPad Software Inc., San Diego, CA, USA). The results were expressed as the mean ± SD. The D’Agostino–Pearson test was used to assess whether the data were normally distributed. Two-way analysis of the variance (ANOVA) was used to compare the masks between the different body positions and in the x , y , and z axes. Variations in the landmarks within the different body positions were evaluated by one-way ANOVA for repeated measurements. The differences between the measurements were tested with Tukey’s post hoc test. Friedman ANOVA for repeated measurements and Dunn’s multiple comparison test were used for those not suited for normal dispersion. A P -value of ≤0.0001 was considered statistically significant. Comparisons of the mean differences between the genders were carried out with the t -test or the Mann–Whitney U -test.
Reproducibility of landmark identification
The 3D scans obtained for each body position acquired from randomly selected subjects (five male and five female) were used in order to test the accuracy of observer landmark identification. The landmarks were labelled and the identifications were repeated two times by two different observers (UO, RS) with a 1-week gap between digitization sessions. The reliability of the observers was assessed using the two-way random intra-class correlation coefficient. A value of 0. or higher was considered to signify reliability.
Intra- and inter-observer reliability of the body positions
The intra-observer and inter-observer reliability for the three body positions was acceptable. The correlation coefficient between the first and second measurements of the sitting position ranged between 0.919 and (average 0.986) for intra-observer reliability, and between 0.914 and 0.999 (average 0.974) for inter-observer reliability. In the standing position, the values ranged between 0.90 and 0.998 (average 0.946) for intra-observer reliability, and between 0.903 and 0.989 (average 0.938) for inter-observer reliability. In the supine position, the results ranged between 0.912 and 0.962 (average 0.946) for intra-observer reliability, and between 0.91 and 0.999 (average 0.965) for inter-observer reliability.
Comparison of facial masks
The similarity of the surfaces was analyzed by comparing the RMS, the MAD, and the MSD.
The results for the merged masks are shown in Fig. 4 . The RMS values between the scans were .001 ± 0.397 for sitting to supine (Sit–Sup), .033 ± 0.408 for standing to supine (Stn–Sup), and .051 ± 0.326 for sitting to standing (Sit–Stn). The results of the MAD were .133 ± 0.362 for Sit–Sup, .2 ± 0.385 for the Stn–Sup, and .165 ± 0.2519 for the Sit–Stn. The MSD values were 0.619 ± 0.267 for Sit–Sup, 0.694 ± 0.278 for Stn–Sup, and −0.073 ± 0.171 for Sit–Stn. Comparisons between the Sit–Sup and the Stn–Sup groups showed no significant differences. The results for Sit–Stn were significantly lower than those of both of the other groups. Comparisons in the x , y , and the z axes were also performed and are shown in Fig. 4 .
Comparison of the landmarks
With further analysis of the data, the mean differences of landmarks in the x , y , and z coordinates were compared among the three body positions. The data are shown in Table 1 and the results are summarized in Table 2 .
|x -axis||y -axis||z -axis|