The application of traditional and geometric morphometric analyses for forensic quantification of sexual dimorphism: preliminary investigations in a Western Australian population.
ABSTRACT A current limitation of forensic practice in Western Australia is a lack of contemporary population-specific standards for biological profiling; this directly relates to the unavailability of documented human skeletal collections. With rapidly advancing technology, however, it is now possible to acquire accurate skeletal measurements from 3D scans contained in medical databases. The purpose of the present study, therefore, is to explore the accuracy of using cranial form to predict sex in adult Australians. Both traditional and geometric morphometric methods are applied to data derived from 3D landmarks acquired in CT-reconstructed crania. The sample comprises multi-detector computed tomography scans of 200 adult individuals; following 3D volume rendering, 46 anatomical landmarks are acquired using OsiriX (version 3.9). Centroid size and shape (first 20 PCs of the Procrustes coordinates) and the inter-landmark (ILD) distances between all possible pairs of landmarks are then calculated. Sex classification effectiveness of the 3D multivariate descriptors of size and shape and selected ILD measurements are assessed and compared; robustness of findings is explored using resampling statistics. Cranial shape and size and the ILD measurements are sexually dimorphic and explain 3.2 to 54.3 % of sample variance; sex classification accuracy is 83.5-88.0 %. Sex estimation using 3D shape appears to have some advantages compared to approaches using size measurements. We have, however, identified a simple and biologically meaningful single non-traditional linear measurement (glabella-zygion) that classifies Western Australian individuals according to sex with a high degree of expected accuracy (87.5-88 %).
- SourceAvailable from: Eugénia Cunha[Show abstract] [Hide abstract]
ABSTRACT: Sex estimation is extremely important in the analysis of human remains as many of the subsequent biological parameters are sex specific (e.g., age at death, stature, and ancestry). When dealing with incomplete or fragmented remains, metric analysis of the tarsal bones of the feet has proven valuable. In this study, the utility of 18 width, length, and height tarsal measurements were assessed for sex-related variation in a Portuguese sample. A total of 300 males and females from the Coimbra Identified Skeletal Collection were used to develop sex prediction models based on statistical and machine learning algorithm such as discriminant function analysis, logistic regression, classification trees, and artificial neural networks. All models were evaluated using 10-fold cross-validation and an independent test sample composed of 60 males and females from the Identified Skeletal Collection of the 21st Century. Results showed that tarsal bone sex-related variation can be easily captured with a high degree of repeatability. A simple tree-based multivariate algorithm involving measurements from the calcaneus, talus, first and third cuneiforms, and cuboid resulted in 88.3 % correct sex estimation both on training and independent test sets. Traditional statistical classifiers such as the discriminant function analysis were outperformed by machine learning techniques. Results obtained show that machine learning algorithm are an important tool the forensic practitioners should consider when developing new standards for sex estimation.Deutsche Zeitschrift für die Gesamte Gerichtliche Medizin 09/2014; · 2.60 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: In the last decade, the advances in imaging technologies have intensified the use of multislice computed tomography (MSCT) for anthropological purposes. The published literature has proved it to be a suitable tool for establishing sexually dimorphic characteristics in different anatomical areas. In this context, the main purpose of the present study was to explore the accuracy of traditional morphometric method when applied to data acquired in three-dimensional (3D) reconstructed os coxae of living Spanish population, to develop a series of statistically robust patterns for sex assessment and to test their validity in innominate remains. For this purpose, 150 volume-rendered innominate CT scans were selected to examine nine interlandmark linear distances by means of descriptive statistics and discriminant function analyses (DFA) employing the jackknife procedure for cross-validations. Five measurements were sexually dimorphic. Acetabular diameter was the single most accurate predictor. This, combined with innominate height and innominate breadth, contributed very significantly to sex estimation. Overall, classification accuracies were 89.3-95.3 % (89.3-94.7 % after cross-validation), with a sex-bias lower than 5 %. The second validation approach performed on a sample of 96 individuals from another contemporary Spanish reference collection comprising dry bones showed the high percentage of accurate classification (83.3-95.8 %). In conclusion, the assessment of sex using cross-sectional MSCT images of the os coxae is possible and the discriminant functions obtained on Spanish living individuals can also be effective for estimating sex from skeletal remains.Deutsche Zeitschrift für die Gesamte Gerichtliche Medizin 06/2014; 128(5). · 2.60 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: A number of previous studies have demonstrated that osteometric analysis of the sternum provides a highly accurate method for discriminating adult sex in diverse population groups. In this study, sternal measurements were recorded from posteroanterior digital radiographs of the chest plate of 116 Spanish individuals (65 males and 51 females). Results demonstrated that all linear dimensions of the manubrium and mesosternum, sternal area, and sternal index were significantly sexually dimorphic in this population group. Discriminant function analyses incorporating several of these variables, individually or in combination, provided sex classification accuracy rates greater than 80.0 %, with associated sex biases below 5.0 %. A stepwise procedure, which can be used when a complete sternum is present, yielded the highest correct sex classification rate at 89.7 %. Only slightly lower allocation accuracy rates were obtained for multivariate equations which incorporated either dimensions of the manubrium or mesosternum (87.1 % for both formulae). Thus, the derived discriminant function equations should prove useful in forensic investigations, particularly those in which the pelvis or bones of the extremities are not available for analysis.Deutsche Zeitschrift für die Gesamte Gerichtliche Medizin 09/2013; 128(2). · 2.69 Impact Factor