Gender and Stature Estimate Based on Facial Measurements: an Anthropometric Study in Southwest Iranian Population

This study was aimed to evaluate the facial dimensions and their relation with gender and stature in the Iranian southwest population. A cross-sectional study was conducted among 300 southwest Iranian cases age 20-50 years (150 males and 150 females). Studied variables through physical anthropometry in both genders were Upper facial height (UFH), Total facial height (TFH), Facial Height (FH), Facial Width (FW), and intercanthal width (IC) using a standard sliding caliper and Collis. Regarding the sex and age of the individuals, general descriptive analysis of facial dimensions was determined. All measurements, except TFH and FH, were different between men and women (p≤ 0.01). In male subjects, the Pearsons correlation coefficient (r) revealed that IC (0.72 ⃰ ⃰ ), FW (0.58 ⃰ ⃰ ) and UFH (0.18 ⃰⃰ ⃰ ) parameters had a positive correlation with stature. In the female group, none of the facial parameters had a significant correlation with stature. Regarding the value of each facial diameter in discriminating male and female stature and gender, the highest discriminative value was specified to the FW (cutoff: 11.89, sensitivity: 89%, specificity: 11%) and IC (cutoff: 2.26, sensitivity: 98%, specificity: 0.04%) respectively. This study showed a significant Brazilian Journal of Forensic Sciences, Medical Law and Bioethics 11(1):22-33 (2021) 23 A. E. Moghadam et al. association of some facial measurements with stature and gender in the southwest Iranian population. IC and FW had the most diagnostic value for gender and stature definition. It is shown a useful reference for the adult southwest Iranian population for facial recognition and that the subject’s sex should be considered during body identification procedures.


Introduction
Anthropometry involves quantitative measurements of muscle, bone, and adipose tissue to evaluate the composition of the body 1 . Anthropometric data are demanding to provide information about body dimension, human profile estimation, Medical anthropology, malnutrition, criminology, and forensic anthropology 2,3 . Nowadays there are very careful techniques to measure standard identification criteria in different populations 4 . Various methods have been evolved for live human identification, including fingerprints, DNA profiling, retina scan, facial features, and many others 5 . But body identification can be achieved based on anthropometric data, skin analysis, dental records, and genetics. The principal elements of anthropometry are gender, height, and age estimate in physical examination 6 . Based on different studies, these elements reveal proportional biological ratios with different body segments which varies between different racial groups 7-10 .
Facial anthropometry is a practical measurement of facial soft and bony tissue for scientific and practical goals: from biometry up to medicine, genetics, pattern recognition, and criminology 11 . This uncomplicated, economical, valid, and non-invasive method effectual for facial morphology analysis and it involves getting direct clinical measurements just as linear distances, dimensions, angles, and ratios 12 . Given the differences based on hereditary, environmental factors, diet quality, and ethnicity between the nation's, performance of specific measurements in each region are necessary to create usable database 13 .
Beside these reasons, we designed this research to elucidate the norms of facial dimensions in the southwest Iranian population to provide database for estimation of gender and stature. Amongst the anthropometrical research methods, morphological and anthropological approaches which rely on the Brazilian Journal of Forensic Sciences, Medical Law and Bioethics 11(1): 22-33 (2021) A. E. Moghadam et al. examinations of the bones and bone measurements respectively, we selected anthropological assay which presents less costly with high-rate validity 3 .

Methods
This prospective cross-sectional study was conducted on the 300 native

Statistical analysis
Data were evaluated in Stata 14 software using independent t-test, chi-square, logistic regression, and receiver operating characteristics (ROC) curve analysis.
The mean distance was reported for the t-test; the odds ratio (OR) for logistic regression; and the area under the curve for the ROC analysis (AUC). The coefficient of regression and 95% confidence intervals (CIs) for each independent variable were measured. p-value < 0.05 was considered as the significance level.

Results
In this study, 50.5% and 49.5% of the participant were male and female respectively. The mean ages of the subjects (male: 32.34 ± 2.03, female: 33.43 ± 2.28) were not significantly different between the two genders. Average facial measurements in the total population base on gender were calculated, the summary of which is illustrated. All measurements, except TFH and FH, were different between both gender (p < 0.01) ( Table 1).   Table 2).
In male subjects IC (0.72 ** ), FW (0.58**), UFH (0.18**) parameters has shown significant correlation with stature ut an i n't t have a significant correlation with stature ( Table 2). In female subjects, none of the facial parameters had a significant correlation with stature (Table 3). The points with significant association were considered for ROC curve analysis. According to this, area under the curve was 94% (p-value=0.000) that has been shown the higher recognition power of FW in stature estimation ( Figure 2).
The area under the curve was 99% (p-value=0.000) that has been shown the higher recognition power of IC in gender estimate (Figure 3).
When the UHF is higher than 5.32 represents the person be a boy and when the UHF is lower than 5.32 represents the person be a girl (Figure 4).

Discussion
The study ahead aimed to estimate the association of gender and stature with facial measurements when incomplete and fragmentary skull materials gleaned during forensic exploration. Concerning having the baseline data for gender, age and stature identify that would be helpful in reduction of identification  In Iran, despite notable research in this field, less attention has been allotted on gender and stature prediction based on facial dimensions in divers diverse races that dwelled in different regions [9][10][11] .
To the best of our knowledge, this is the first study to assess facial dimensions in the southwest of Iran. Average facial measurements in the total population base on gender were calculated, all measurements, except TFH and FH, were different between males and females. Effects of genetics, age, nutrition, and sex hormones on facial features during maturation could explain these differences 12

Conclusion
The present study findings indicated that measuring facial parameters in the studied samples could accurately estimate gender and stature.