Towards a Unified Facial Tissue Thickness Model from Neuroimaging Data for Selected Age Groups of Sri Lankan Adult Population

Abstract

Abstract Facial soft tissue thickness (FSTT) is essential in forensic anthropology for facial reconstruction, recreating a recognizable face from an unidentified skull and plastic surgeons for treatment planning. Together with the age and sex of a person, the facial profile is related to facial soft tissue thickness, which is required for accurate facial reconstruction and recognition. In this study facial soft tissue thickness was measured at 23 standard anthropological landmarks using Magnetic Resonance Images (MRI) of 243 adult individuals (male - 123, female - 120) of the Sri Lankan adult population which were collected as clinical data from the National Hospital of Sri Lanka. Inter observer variability of the measurements were measured and it was proved that the difference between measurements of the three observers were minimum. For each landmark, average thickness value, standard deviation, range, maximum and minimum facial soft tissue thickness values classified according to the gender and age were documented. Mean FSTT classified according to age and gender was then compared with age groups, gender groups and studies of three foreign countries selected. All the comparisons were done based on graphical visualizations and statistical methods. Statistical analysis and learning methods were used for missing FSTT value imputation individual cases. From the comparison of FSTT within gender groups it was found that men have comparatively higher FSTT than women. The area along the midline of men always shows higher FSTT than female. The area around the cheeks (represented by Jugale, Zygomatic Arc and Supra Glenoid) also show comparatively large tissue thickness in young men (within 20-39 age range) than in women groups. From the comparison of FSTT within age groups it was observed that the thickness at some landmarks decrease or increase with age while some do not show a specific formal pattern variation with aging. Also it was found that males have the highest FSTT at the age range of 40-49. In both males and females FSTTs at points like Midphiltrum and Upper Lip Margin gradually decrease with the age. In females the FSTTs at Supra Orbital, Jugale, Zygomatic arch and Supra Glenoid landmark increase with age. Males have the least FSTT mean for Supra Orbital, Jugale, Zygomatic arch and Supra glenoid landmarks at the age range 20-29, but it doesn‘t show a direct or inverse relationship between FSTT and age when other age ranges are considered. With the comparison of three foreign studies (Taiwanese, Turkish and North West Indian) with the Sri Lankan study it was observed that Taiwanese population shows lesssignificant difference with compared to the Sri Lankan population. However due to the unavailability of raw data statistical interpolation of missing values using foreign data was not performed. A part from statistical analysis learning methods were established towards estimating missing data values. A regression tree analysis was done for 40-49 and 50-59 age groups with the clustered components(which comprised of landmarks) using Principal Component Analysis and Factor Analysis methods. Hence the objective of interpolation of missing values towards making a unified facial tissue thickness model was achieved through this.