Complications in 3D Quantitative Analysis of Sexual Dimorphism of the Human Nuchal Area

Established visual scoring guidelines identify the mastoid process, supraorbital margin, browridge, and the nuchal crest as standard sexually dimorphic features of human crania. Recent work has utilized 3D surface scanning to quantify sexual dimorphism in the browridge and mastoid process, 2 of the 4 cranial features most commonly used to determine sex following visual criteria. We aimed to establish a method to measure sex differences in nuchal area volume from 3D surface scans. A pilot study was performed using a sample (N = 20) from a Medieval Nubian population for which sex determinations were available based on the bony pelvis and in some cases preserved soft tissue. A Polhemus FastScan 3D laser surface scanner was used to capture cranial surface morphology. For each specimen, the superior and basicranial mesh models were imported, aligned and merged in Geomagic Control 2015 to create a single 3D model of the cranium. The nuchal area was defined using two landmark‐based approaches to guide placement of a plane: 1) lambda to opisthion (L‐O Nuchal Vol), and 2) a more targeted analysis using inion to opisthion (I‐O Nuchal Vol), and total cranial volume was measured from the models using Geomagic's volume to plane computation tool which was also utilized to measure nuchal volumes. Then, a nuchal:cranial volume ratio was computed, and t‐tests were used to compare sex specific means. Total cranial volume was significantly different ( p = 0.0011) between males and females (male average = 1884478mm 3 ± 153537.6; female = 1670487mm 3 ± 85547.83), however, raw nuchal area volume using the lambda to opisthion plane (male average = 108261.8mm 3 ± 26517.98; female = 91768.95mm 3 ± 19077.56) was not ( p > 0.05). The L‐O nuchal volume ratio results were also not significantly different ( p > 0.05) between the sexes (male average = 0.057 ±0.01; female = 0.055 ±0.01). Similarly, non‐significant results were obtained using the more targeted I‐O nuchal volume ratio. Despite clear differences in the surface appearance of male versus female nuchal areas in the study population, with males exhibiting more pronounced nuchal lines and general increased rugosity across the nuchal region compared to females, the results of this pilot study did not identify significant differences between the sexes in raw or scaled nuchal area volume. We examined potential sources of confounding variables and methodological issues. Possible confounding factors include a small sample size, a highly variable sample population (L‐O Nuchal Vol: CV = 22.47 in males, 22.69 in females; I‐O Nuchal Vol: CV =, 43.39 in males, 35.71 in females), as well as overall shape differences between male and female neurocrania. Methodological issues regarding plane placement may also bias the results. Nuchal volumes were collected using a built‐in tool to calculate volumes above and below a plane placed in an object; hence, plane position is highly dependent on landmark location, which cannot be saved with the model in the software. This may lead to repeatability issues when the plane is not placed correctly. Loss of surface detail in rendered scans may also alter the results as landmark placement is dependent on surface detail. Future work is underway to resolve these issues and to establish a repeatable and reliable protocol to quantify the nuchal area based on 3D surface scans.