L’âge est un paramètre du profil biologique des individus immatures que l’on peut estimer de manière suffisamment fiable et précise. Les méthodes utilisées ne respectent pas toutes des critères méthodologiques valides et sont donc relativement critiquables. Une analyse de 256 méthodes utilisées en anthropologie biologique a qualifié et quantifié les biais et limites de construction et d’application et propose une classification objective des méthodes mettant en évidence celles qui sont valides. Cette analyse a servi à élaborer le protocole d’une nouvelle méthode standardisée. Les surfaces de l’os iliaque, la cinquième vertèbre lombaire et la clavicule ont été reconstruites à partir de CT scans avec le logiciel AVIZO®. Des variables biométriques et non-biométriques ont été obtenues sur ces trois os. Des équations non-paramétriques de type Multivariate Adaptive Regression Splines (MARS) prédisant l’âge à partir des variables ont été calculées avec le logiciel statistique R®. Les variables iliaques, les hauteurs latérales vertébrales et la longueur maximale de la clavicule ont donné les meilleures prédictions de l’âge. Aucune différence significative entre les âges réels et estimés n'a été observée sur trois échantillons-test de Marseille, Toulouse et la collection ostéologique Luis Lopes. Les résultats sont meilleurs ou comparables à ceux d’autres méthodes. Des probabilités d’âge en fonction des indicateurs non-biométriques ont donné des résultats suffisamment fiables, mais pas assez précis pour être valides. Notre méthode d’estimation de l’âge a 95% de fiabilité et 90% de justesse. Elle est statistiquement valide et utilisable en contexte médico-légal et archéologique. / Age is one of the parameters of a juvenile biological profile that can be estimated with sufficient reliability and precision. Many juvenile age estimation methods do not follow valid methodological criteria. A critical analysis of 256 methods used in physical anthropology was done to qualify and quantify biases and limitations of method construction and application. An objective and practical classification of methods highlighting those respecting valid sampling protocols and statistical criteria was constructed. This analysis provided a protocol for a new standardised juvenile age estimation method. Surfaces of the iliac bone, fifth lumbar vertebra and clavicle were reconstructed from CT scans of individuals from Marseilles using the AVIZO® software. Biometric growth variables and non-biometric maturation indicators were taken on the three bones. Non-parametric Multivariate Adaptive Regression Splines (MARS) models were calculated with the R® software to predict age using the variables. The best models were obtained with variables of the ilium, left and right vertebral heights and maximal clavicular length. These models were validated on three samples from Marseilles, Toulouse and the Luis Lopes osteological collection. They were comparable or better than other age estimation methods. Posterior probabilities of age were calculated using maturation stages of four iliac epiphyses. Results were reliable, but precision was too low for validation. Our new juvenile age estimation method can predict age with 95% reliability and 90% accuracy. It is standardised, statistically valid, integrates changes in variability and is applicable in a forensic or bioarchaeological context.

Histological methods to estimate skeletal age at death, in forensic cases, are an alternative to the more traditional gross morphological methods. Most histological methods utilize counts of bone type within a given field for their estimation. The method presented in this paper uses the percentage area occupied by unremodeled bone to estimate age. The percentage area occupied by unremodeled bone is used in a linear regression model to predict skeletal age at death. Additionally, this method uses digital software to measure area rather than the traditional technique in which a gridded microscope is used to estimate area. The clavicle was chosen as a sample site since it is not a weight bearing bone and has little muscular insertion. These factors reduce the variation seen as a result of differences in lifestyle or activity pattern.

Transition analysis transforms skeletal traits with an invariant, unidirectional series of stages into a likelihood function with a maximum likelihood value and a 95% confidence interval. Boldsen et al. used transition analysis to develop an adult age estimation method employing components of the cranial sutures, pubic symphysis and ilial portion of the sacroiliac joint, used either in combination or individually. This validation study aimed to use the 36 transition analysis numerical, categorical scores for the anatomical features in conjunction with the ADBOU computer program to assess the accuracy and precision of the age estimates for 149 black individuals from the Pretoria Bone Collection. In addition, the effect of observer variability in scoring of these traits was assessed. Six age estimations were generated by the ADBOU computer program using 1) the cranial sutures only, 2) the pubic symphysis only, 3) the auricular surface of the ilium only, 4) all three features combined, 5) all three features combined and modified by a forensic prior distribution and 6) all three features combined and modified by an archaeological prior distribution. The six point estimate categories, calculated from the maximum likelihood values, were evaluated for accuracy using mean absolute values. The 95% confidence intervals were evaluated for range width and accuracy. Cohen’s Kappa statistics were used to analyse repeatability of the scoring procedure through inter- and intra-observer agreement and Kruskal-Wallis ANOVA statistics to determine the effect of observer differences on the final age estimates. The usefulness of the age ranges were diminished by large widths encompassing up to 95 years. The accuracy for the point estimates fared better for the combined skeletal indicators and overall accuracy was improved by using the archaeological prior distribution. The archaeological prior distribution was also responsible for narrowing the age ranges, especially in the older ages (over 70 years). Age estimates did not differ significantly when using inter- and intra-observer scores, but experience with the method did seem to improve results. Overall, age ranges were too wide, but accuracy could potentially be improved by adding more skeletal components to the method and using a population-specific prior distribution. The method would need considerable adjustments to make it usable in a South African setting. / Dissertation (MSc)--University of Pretoria, 2014. / tm2015 / Anatomy / MSc / Unrestricted

Big sagebrush (Artemisia tridentata Nutt.) is the most widespread and common shrub in the sagebrush biome of western North America. Of the three most common subspecies of big sagebrush (Artemisia tridentata), mountain big sagebrush (ssp. vaseyana; MBS) is the most resilient to disturbance, but still requires favorable climactic conditions and a viable post-fire seedbank for successful unassisted recovery. This study was designed to assess MBS seed production throughout post-fire recovery. We performed 2 pilot studies to develop methods for estimating seed production and plant age. The results of the pilot studies and a space-for-time substitution strategy were used to measure seed production on 13 sites ranging from 10-33 years post-fire. We hypothesized that seed rain (mean seeds produced/ m2) would peak before stand density had maximized due to decreasing individual plant fecundity (mean seeds produced/ plant) in high density stands. We measured population density and individual plant fecundity for three size classes of MBS and used forward stepwise regression analysis to identify environmental factors influencing seed production over time. Density for small (basal stem diameter (BSD) <1 cm) and medium-sized (BSD=1-3 cm) plants was consistently low and was not affected by time since fire (TSF), while large-sized (BSD > 3 cm) plant density increased steadily with TSF (p=0.0002). Plant fecundity decreased with TSF for all three size classes (p range = 0.019 – 0.0506), with large plants dominating reproductive output. Small and medium-sized plant fecundity was negatively correlated with winter precipitation (p range = 0.0106-0.0174), while large plant fecundity was positively correlated with winter precipitation (p<0.0001) and negatively correlated with elevation (p=0.0001). Despite losses in plant fecundity over time for all size classes, steady recruitment in population density resulted in increased seed rain (p=0.0039), suggesting that increases in stand density compensated for losses in individual plant fecundity. Results partially support our hypothesis that the time required for MBS seed rain to be maximized was not tightly bound to indicators of stand maturation. Understanding the factors that influence post-fire seed production can help land managers better manage for successful recovery by providing them with tools for evaluating seed production capabilities of MBS communities.

Accurate estimates of parameters and demographic rates of wildlife populations are crucial to management and conservation. One parameter of interest is age because it influences multiple demographic rates and behaviors including reproduction, dispersal, migration, and disease transmission. There are a number of methods used to estimate ages of wildlife. One of the most common methods used for mammals is evaluation of patterns in dental eruption and tooth wear. In the first chapter, we evaluate the accuracy associated with estimates of age from dental eruption and patterns of tooth wear for mule deer (Odocoileus hemionus). In the second chapter, we use age estimates validated within the first chapter to explore how age and other maternal factors may influence number of offspring born per litter in mule deer. We further test relative support for two common hypotheses--terminal investment and reproductive restraint--regarding the relationship between age and reproductive investment in this species. Our results will inform conservation, management, and research efforts associated with mule deer.

Yes / Recent advances in the methods of skeletal age estimation have rekindled interest in their applicability to paleodemography. The current study contributes to the discussion by applying several long established as well as recently developed or refined aging methods to a subsample of 121 adult skeletons from the early medieval cemetery of Lauchheim. The skeletal remains were analyzed by 13 independent observers using a variety of aging techniques (complex method and other multimethod approaches, Transition Analysis, cranial suture closure, auricular surface method, osteon density method, tooth root translucency measurement, and tooth cementum annulation counting). The age ranges and mean age estimations were compared and results indicate that all methods showed smaller age ranges for the younger individuals, but broader age ranges for the older age groups. / Max Planck Institute for Demographic Research

yes / Context: Adult age-at-death is presented in a number of different ways by anthropologists. Ordinal categories predominate in osteoarchaeology, but do not reflect individual variation in ageing, with too many adults being classified as ‘middle adults’. In addition, mean ages (derived from reference samples) are overly-relied upon when developing and testing methods. In both cases, ‘age mimicry’ is not adequately accounted for. Objectives: To highlight the many inherent biases created when developing, testing and applying age-estimation methods without fully considering the impact of ‘age mimicry’ and individual variation. Methods: The paper draws on previously published research (Web of Science, Pub Med, Google Scholar) on age estimation methods and their use in anthropology. Results and Conclusions: There is a lack of consistency in the methods used to estimate age, and for the mode of combining them. Ordinal categories are frequently used in osteoarchaeology, whereas forensic anthropologists are more likely to produce case-specific age ranges. Mean ages reflect the age structure of reference samples, and should not be used to estimate age for individuals from populations with a different age-at-death structure. Individual-specific age ranges and/or probability densities should be used to report individual age. Further research should be undertaken on how to create unbiased, combined method age estimates.

yes / We analyzed the trabecular bone changes in the pubic bone (PB) and in the auricular surface (AS) of the ilium using 319 CT scans of cadavers to estimate the age. Although the sharpness of the trabecular structure decreases in CT images when soft tissues are present, we identified four phases for the changes in PB and five in AS; a juvenile trait in PB and a senile trait in AS helped narrow the age range. High correlation with age was identified for both sexes in PB (F 0.89; M 0.75) and in AS (F 0.85; M 0.71) used independently or combined (F 0.91; M 0.78). The old adults (>60 years) could be evaluated with better accuracy and discriminated in several phases. We found low inter-observer error and low inaccuracy (about 6 years, mean for all age ranges). The method is robust with respect to slice thickness, display window and kernel within the tested ranges.

Yes / The skeletal remains of infants and children are a poignant reminder of the perilous nature of childhood in the past, yet they offer valuable insight into the life histories of individuals and into the health of populations. Many osteoarchaeological and bioarchaeological analyses are dependent on two vital pieces of information: the age-at-death and sex of the individual(s) under study. This chapter will outline how age-at-death and sex can be estimated from the skeletal remains and dental development of non-adults, and how these are easier or more difficult to determine than for adults, and will discuss the complexities and controversies surrounding different methods.

No / Research in facial age estimation has been active for over a decade. This is due to its numerous applications. Recently, convolutional neural networks (CNNs) have been used in an attempt to solve this age old problem. For this purpose, researchers have proposed various CNN architectures. Unfortunately, most of the proposed techniques have been based on relatively ‘shallow’ networks. In this work, we leverage the capability of an off-the-shelf deep CNN model, namely the VGG-Face model, which has been trained on millions of face images. Interestingly, despite being a simple approach, features extracted from the VGG-Face model, when reduced and fed into linear regressors, outperform most of the state-of-the-art CNNs. e.g. on both FGNET-AD and Morph II benchmark databases. Furthermore, contrary to using the last fully connected (FC) layer of the trained model, we evaluate the activations from different layers of the architecture. In fact, our experiments show that generic features learnt from intermediate layer activations carry more ageing information than the FC layers.