Premium
Quantitative Sex Estimation Based on Cranial Traits Using R Functions
Author(s) -
Nikita Efthymia
Publication year - 2019
Publication title -
journal of forensic sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.715
H-Index - 96
eISSN - 1556-4029
pISSN - 0022-1198
DOI - 10.1111/1556-4029.13833
Subject(s) - discriminant , linear discriminant analysis , glabella , mathematics , mastoid process , statistics , forensic anthropology , artificial intelligence , sample (material) , discriminant function analysis , pattern recognition (psychology) , range (aeronautics) , computer science , enhanced data rates for gsm evolution , anatomy , biology , geography , engineering , forehead , archaeology , chemistry , chromatography , aerospace engineering
This paper presents an R script that quantifies the shape of selected cranial traits and automates sex estimation. The proposed functions were tested on two modern Greek assemblages. The discriminant variables input in the functions are calculated from a digital photograph of the lateral view of the cranium. The cranial outline is determined using the Canny edge detector and discriminant variables that quantify the shape of the glabella/frontal bone, mastoid process, and external occipital protuberance are computed. The best cross‐validated results for pooled sexes in the Athens Collection range from 84.2% to 87.3%, and increase up to 93.9% when half of the sample is used for training and the rest for prediction, while correct classification for the Cretan material is 80–90% for optimum combinations of discriminant variables. The greatest advantage of the proposed method is its straightforward and time‐efficient application.