Tracing asymmetry and sexual dimorphism in the adult iliac auricular surface: a geometric morphometrics and machine learning approach.

The iliac auricular surfaces (AuSs) are critical anatomical landmarks in evolutionary research, clinical evaluation, and biological profiling. As paired structures, however, they may exhibit asymmetry, a phenomenon documented in non-adult specimens but less well understood in adults. This study examines whether such asymmetry persists into adulthood by analyzing 100 adult skeletons (25–97 years; 50 males, 50 females) from the 21st Century Identified Skeletal Collection (XXI/CEI, University of Coimbra). High-resolution 2D images of the AuSs were captured using a digital camera, and the joint contours were recorded with landmarks and semilandmarks. These data were analyzed using an integrated framework combining 2D geometric morphometrics, machine learning, and statistical methods. The results reveal no significant asymmetry between left and right AuSs but demonstrate clear patterns of sexual dimorphism. Machine learning algorithms achieved a 70% accuracy rate in sex estimation. This exploratory study underscores the potential of combining morphometric and computational approaches for biological profiling, while highlighting the need for further research to address current limitations.