Morphometrics of the juvenile skull: analysis of ontogenetic growth patterns

This study assesses patterns of ontogenetic growth, and by association biological size and shape variation, in juvenile crania from a Western Australian population. Current methods of juvenile age estimation largely relate to quantifying dental development and skeletal growth. The current study develops age prediction models using linear measurements of the juvenile cranium. The specific aims of the project are threefold: i) to quantify the relationship between cranial form and chronological age in sub-adults using both traditional and geometric morphometric methods; ii) to statistically quantify the timing of cranial fontanelle closure; and iii) to quantify at what age the human cranium is sexually dimorphic.

The study sample comprised 174 cranial multi-slice computed tomography (MSCT) scans of individuals ranging in age from birth to 18.76 years (98 males and 76 females). The MSCT scans are visualized using OsiriX®, 52 homologous landmarks are acquired in each cranial scan. MorphDB and morphologika are then used to calculate linear measurements and view three-dimensional landmark configurations respectively. Anterior fontanelle size was calculated using two methods: finding the area of the open anterior fontanelle using a polygon tracing tool in OsiriX®, and by taking oblique linear distances of the open fontanelle. Prior to collection of all data intra-observer precision levels were quantified.

Age prediction models were developed based on interlandmark distances. Individual and pooled sex models were produced, with prediction accuracy ranging from ±1.3 to ±2.7 years. The single most accurate predictor of age in the pooled sex sample was palatine height (±2.3 years).

Ontogenetic growth patterns were quantified in three cranial regions (neurocranium, viscerocranium and basicranium). It was demonstrated that at birth the neurocranium is proportionally the largest region of the skull, and that the viscerocranium has the longest and greatest amount of growth. It was also demonstrated that at nine years of age the cranium is approximately 93% that of its final adult size. Cranial shape is analysed independent of size. Ontogenetic shape differences related to age are present only on the one axis of variation (PC1): as age increases, the cranium shows an elongation of the viscerocranium, increased anterior projection of the mastoid process, and a widening of the neurocranium.

Assessment of the timing of anterior fontanelle closure demonstrated that by 2.0 years of age, that fontanelle is fully closed in approximately 50% of the Western Australian population examined. By 2.5 years of age, however, 100% of individuals presented complete closure of the anterior fontanelle.

Sexual dimorphism was quantified using basic descriptive statistics and discriminant function analyses of the linear measurements to assess classification accuracy. It was demonstrated that sex could be estimated to an acceptable degree of accuracy (77.6%; sex bias -1.9%) in a post-pubertal sample (i.e. over nine years of age) based on two linear measurements: anterior basicranial breadth, and frontal chord.

The results of the present research have the potential to influence forensic investigations specifically in Western Australia and Australia generally. It has been demonstrated the population specific models can be used to estimate age with an accuracy of ±1.3 to ±2.7 years. Additionally, sex can be accurately estimated in the juvenile cranium from the age of nine years onwards.