Testing the accuracy of the DRNNAGE software for age estimation in a modern Greek sample.

Estimation of age-at-death from human skeletal remains is fundamental in forensic anthropology as part of the construction of the biological profile of the individual under study. At the same time, skeletal age-at-death estimation in adults is problematic due to the disparity between chronological and biological age, the important inter-individual variability at the rate of skeletal aging, and inherent biases in the available methodologies (e.g., age mimicry). A recent paper proposed a method for skeletal age-at-death estimation based on multiple anatomical traits and machine learning. A software was also created, DRNNAGE, for the easy implementation of this method. The authors of that study supported that their methods have very high repeatability and reproducibility, and the mean absolute error of the age estimation was ~6 years across the entire adult age span, which is particularly high and promising. This paper tests the proposed methodology on a modern documented Greek sample of 219 adult individuals from the Athens Collection, with age-at-death from 19 to 99 years old. The sample was split into males and females as well as into individuals under and over 50 years old. We also divided the sample in 10-year intervals. First, intra- and inter-observer error was estimated in order to assess repeatability and reproducibility of the variables employed for age-at-death estimation. Then, the validity (correct classification performance) of DRNNAGE for each anatomical region individually, as well as all combined, was evaluated on each demographic separately and on the pooled sample. According to the results, some of the variables showed very low repeatability and reproducibility, thus their use should be cautious. The DRNNAGE software showed overall highly accurate age-at-death estimates for individuals older than 50 years, but poor on younger adults, with only exception the cranial sutures, which performed surprisingly well for all age groups. Overall, these results support the importance of cross-validation and the use of population-specific methods in forensic anthropology.

Dental diseases and dental wear as a proxy for dietary patterns in Hellenistic-early Roman Menainon, Sicily.

OBJECTIVE: This paper explores dental diseases and wear as a proxy for dietary patterns in Hellenistic-early Roman Menainon. MATERIALS: This study includes 166 individuals (4th-1st c. BCE). METHODS: Carious lesions, dental calculus, antemortem tooth loss, and dental wear were recorded to explore male-female and adult-juvenile differences, and to position Menainon in the broader Hellenistic/early Roman world through comparisons with published data from other sites. RESULTS: Males and females showed similar rates of dental diseases. Dental wear, in contrast, was systematically greater in males. Caries rates were high in both adults and juveniles, but adults showed more calculus. The population from Menainon had higher frequencies for calculus and carious lesions compared to contemporary Italian and Greek assemblages, and a similar frequency for antemortem tooth loss. CONCLUSION: Some sex-related differences in the dietary patterns of the Menainon population were visible but small. The diet of adults and juveniles must have been similar in terms of carbohydrate consumption but different with regard to protein consumption. The high frequency of carious lesions and calculus compared to other Greco-Roman sites suggests that this population must have had good access to dietary resources (protein and carbohydrates). SIGNIFICANCE: This paper provides insights on gender (sex-related) and age divisions in the Hellenistic/early Roman society through the exploration of food consumption in a Sicilian assemblage. LIMITATIONS: Dividing the assemblage by sex and age group reduced considerably the sample size. SUGGESTIONS FOR FURTHER RESEARCH: Stable isotopes and dental microwear analyses should be used to investigate dietary patterns further.

Mechanical stress in the urbanized Roman Phoenician coast.

OBJECTIVE: Skeletal populations from Byblos, Beirut, and Tyre were studied to assess mechanical stress along the Roman Phoenician coast. MATERIALS: The sample included 153 adult skeletons. METHODS: Skeletal remains were macroscopically assessed for osteoarthritis, intervertebral disc disease (IDD), and Schmorl's nodes. RESULTS: The Byblos population experienced higher levels of mechanical stress than the Beirut and Tyre ones. Sex-based differences were also found in all skeletal assemblages with males likely engaging in physically more demanding tasks. CONCLUSIONS: The variation in mechanical stress, and associated physically demanding tasks, between these populations can be attributed to their differing political and economic status during the Roman period; textual sources highlight the economic and political dominance of Beirut and Tyre, emanating from their status as coloniae. SIGNIFICANCE: This study represents one of the first attempts to investigate mechanical stress in coastal Phoenicia during the Roman period. It provides valuable insights into the biocultural structure of understudied communities at the periphery of the Roman world, and can serve as a basis for further future research into the occupational patterns of Phoenician communities. LIMITATIONS: The contextual information for these skeletal populations is very limited and does not allow secure conclusions regarding their representativeness. The sample sizes are also rather small, especially when divided per sex and age. SUGGESTIONS FOR FURTHER RESEARCH: Further investigation employing complementary methods such as cross-sectional geometric properties and entheseal changes is needed to reconstruct the occupational patterns of these communities, taking into account cultural, environmental, and temporal factors.

Beyond dirty teeth: Integrating dental calculus studies with osteoarchaeological parameters.

The study of ancient human dental calculus (mineralized dental plaque, also known as tartar) is becoming increasingly important in osteoarchaeology, human palaeoecology and environmental archaeology. Microremains of different origin (e.g. starch granules, pollen, phytoliths, feather barbules) as well as biomolecules and chemical compounds retrieved from its mineral matrix may represent an important link between past humans and their physical, biological and social environment, but they are rarely fully linked to the evidence from skeletal remains. This paper critically reviews the lines of evidence retrieved from dental calculus in relation to osteoarchaeological parameters, employing macroscopic, microscopic and biomolecular approaches, assessing synergy potential and limitations. The scope of this paper is also to contribute to the building of a much needed theoretical framework in this emerging subfield.

Isotopic evidence for human mobility in late antique Bulla Regia (Tunisia).

This paper represents the first isotopic study on late antique human mobility in North Africa, using the urban site of Bulla Regia in Tunisia as a case study. We also present the first values for bioavailable 87Sr/86Sr in northern Tunisia, analysing 63 plant and snail samples, as well as a simple method for the pre-processing of plants in the field to facilitate their export. Bulla Regia was a prominent Roman and late antique town situated on an important axis of transport and communication in North Africa and is therefore an ideal site to explore mobility in the region during this time period. Strontium (87Sr/86Sr) and oxygen (δ18OCarb) isotopic analysis of 22 late antique individuals from a Christian church and cemetery identified at least seven or eight non-locals, while comparative analysis of five Roman individuals from a funerary enclosure on the same site classified all but one of them as potential locals. Most non-local individuals exhibit 87Sr/86Sr values that match various areas of northern Tunisia, which supports regional mobility rather than long-distance migration, although when combined with the oxygen results, inter-regional mobility from an area with a warmer climate may be hypothesised for some individuals. Examination of the spatial distribution of non-local individuals in their cemetery setting reveals that they were privileged individuals, thus they may reflect the mobility of wealthier town-dwellers in late antiquity, particularly perhaps along the Carthage-Hippo route.

Nutrition, obesity, and dental development in young adolescents in Chicago.

OBJECTIVES: Childhood obesity is a systemic disease with multiple downstream consequences, including shifts in timing of growth and development. It has been documented that children with high body mass index (BMI) show accelerated timing of dental development, but the mechanism for this acceleration is unknown. Prior work has suggested that inflammation and/or nutrition may play a role. We investigate the potential association between diet (caloric intake, macronutrients), obesity, and accelerated dental development. METHODS: Children and adolescents (age 10-15; n = 112) were recruited from dental clinics at the University of Illinois Chicago. We collected subjects' height, weight, panoramic radiographic records, and each subject filled out a Block Food Frequency Questionnaire. RESULTS: The only macronutrient level associated with BMI was a negative correlation to Total Fat consumption (p = .01), though this relationship was not significant in the path analysis (p > .05). Regression analyses indicated that BMI (p = .003) and total caloric intake (controlling for BMI; rho = 0.19; p = .04) were both significantly correlated with timing of dental development. However, when a path analysis was conducted, it was revealed that only BMI was statistically significant (p = .008). CONCLUSIONS: Body mass index percentile, regardless of caloric intake, is positively associated with accelerated dental development. While it is possible that excess caloric intake itself plays a minor role in timing of dental development, we do not see unambiguous evidence for this in our sample. We posit that another mechanism, such as inflammation, may be the link between obesity status and dental development.

On the use of machine learning algorithms in forensic anthropology.

The classification performance of the statistical methods binary logistic regression (BLR), multinomial and penalized multinomial logistic regression (MLR, pMLR), linear discriminant analysis (LDA), and the machine learning algorithms naïve Bayes classification (NBC), decision trees (DT), random forest (RF), artificial neural networks (ANN), support vector machines (linear, polynomial or radial) (SVM), multivariate adaptive regression splines (MARS), and extreme gradient boosting (XGB) is examined in skeletal sex/ancestry estimation. The datasets used to test the performance of these methods were obtained from a documented human skeletal collection, Athens Collection, and the Howells Craniometric data set. For their implementation, an R package has been written to search for the optimum tuning parameters under cross-validation and perform sex/ancestry classification. It was found that the classification performance may vary significantly depending on the problem. From the methods tested, LDA and the machine learning technique of linear SVM exhibit the best performance, with high prediction accuracy and relatively low bias in most of the tests. ANN and pMLR can generally be considered to give satisfactory predictions, whereas NBC when using metric traits and DT are the worst of the classification methods examined. The possibility of making the models developed via the machine learning algorithms applicable to other assemblages without the use of a training sample is also discussed.

Estimation of adult age-at-death from entheseal robusticity: A test using an identified Italian skeletal collection.

OBJECTIVES: The present study tests the effectiveness of entheseal robusticity (ER) as a potential predictor of adult age-at-death by applying multiple regression models to a large contemporary identified skeletal sample. MATERIALS AND METHODS: ER was recorded for 23 bilateral postcranial entheses on 481 adult individuals (271 females and 210 males) from the Frassetto identified skeletal collection of Sassari (Italy), following the method of Mariotti et al. 2007. ER scores were used as predictors in multiple regression analyses with age as the dependent variable; the performance of the resulting models was tested through standard error of estimate, the correlation coefficient between predicted and documented age, and the percentage of correctly classified individuals. RESULTS: ER show a higher correlation with age in females, resulting in more accurate estimates when compared with those obtained for males. Age-at-death is overestimated for male individuals under 45 years old and underestimated for older individuals. Regression models including a reduced set of ER sites result in better estimates compared with those using the complete set of variables. DISCUSSION: Results suggest the potential usefulness of ER for the estimation of age, especially for female skeletal remains. If used with caution, ER may be a valid complement to existing methods for the reconstruction of the biological profile of skeletal remains of archaeological and forensic interest.

Sex estimation: a comparison of techniques based on binary logistic, probit and cumulative probit regression, linear and quadratic discriminant analysis, neural networks, and naïve Bayes classification using ordinal variables.

The performance of seven classification methods, binary logistic (BLR), probit (PR) and cumulative probit (CPR) regression, linear (LDA) and quadratic (QDA) discriminant analysis, artificial neural networks (ANN), and naïve Bayes classification (NBC), is examined in skeletal sex estimation. These methods were tested using cranial and pelvic sexually dimorphic traits recorded on a modern documented collection, the Athens Collection. For their implementation, an R package has been written to perform cross-validated (CV) sex classification and give the discriminant function of each of the methods studied. A simple algorithm that combines two discriminant functions is also proposed. It was found that the differences in the classification performance between BLR, PR, CPR, LDA, QDA, ANN, and NBC are overall small. However, LDA is simpler and more flexible than CPR, QDA, and ANN and has a small but clear advantage over BLR, NBC, and PR. Consequently, LDA may be preferred in skeletal sex estimation. Finally, it is striking that the combination of pelvic and cranial traits via their discriminant functions, determined either by BLR or LDA, removes practically any population-specificity and yields much better predictions than the individual functions; in fact, the prediction accuracy increases above 97%.

Population history of southern Italy during Greek colonization inferred from dental remains.

OBJECTIVES: We are testing competing scenarios regarding the population history of the ancient Greek colonization of southern Italy using dental phenotypic evidence. MATERIALS AND METHODS: We collected dental metric and nonmetric trait data for 481 human skeletons from six archaeological sites along the Gulf of Taranto, dating to pre-colonial (900-700 BC) and post-colonial periods (700-200 BC). We are evaluating scenarios through an individual-level biodistance analysis using a three-pronged approach: (a) by analyzing levels of mobility in pre- and post-colonial periods under a model of isolation-by-distance; (b) by quantifying differences in group means and variances in pre- and post-colonial periods utilizing permutational multivariate analysis of variance and Betadisper analyses; and (c) by identifying ancestries of post-colonial individuals using naïve Bayes classification. RESULTS: Southern Italy during pre-colonial times was characterized by low levels of mobility and marked differences in group means and variances. During post-colonial times, mobility increased and there were no differences in group means and variances. About 18% of the people in post-colonial times were of Greek ancestry and lived equally distributed across Greek colonies and indigenous villages. Nevertheless, the overall biological composition and variability of southern Italy remained relatively unchanged across pre- and post-colonial periods. DISCUSSION: Our results support a scenario in which only few Greek colonists migrated to southern Italy and lived in smaller numbers alongside indigenous people in Greek colonies as well as in indigenous villages. Our results contradict a scenario in which large numbers of Greek invaders founded biologically isolated and substantially homogeneous colonial enclaves within conquered territories.

A three-dimensional digital microscopic investigation of entheseal changes as skeletal activity markers.

OBJECTIVES: The current article explores the effectiveness of entheseal changes (EC) as skeletal activity markers by testing the correlation between such changes and cross-sectional geometric (CSG) properties while controlling for the effect of age and body size. MATERIALS AND METHODS: The originality of the article lies in capturing EC in a continuous quantitative manner using three-dimensional microscopy. Roughness and bone resorption were recorded on Zones 1 and 2 of three humeral entheses (subscapularis, supraspinatus, and infraspinatus) in a documented sample of 29 male skeletons. RESULTS: Our analysis found that merely 5.91% of the partial correlations between EC and CSG properties were statistically significant. In addition, two unexpected patterns were identified, namely a higher number of significant correlations on the left side entheses compared to the right side ones, and a higher number of correlations between minimum roughness and CSG properties compared to mean and maximum roughness. DISCUSSION: These patterns are the inverse of what we would expect if activity had exerted an important effect on EC expression. Therefore, they support the lack of association between EC and habitual activity, even though various factors potentially affecting the above results are discussed.

Skeletal age-at-death estimation: Bayesian versus regression methods.

Age-at-death estimation in a skeletal assemblage (target sample) is biased by the demographic profile of the material used for age prediction (training sample) when this profile is different from that of the target sample. This bias is minimized if the demographic profile of the target sample is properly taken into account in the method developed for age-at-death estimation. In the Bayesian approach this is accomplished via the informative prior. For methods based on regression, we propose two techniques: (a) using weighting factors taken from the demographic profile of the target sample, and (b) creating a new hypothetical training sample that has a demographic profile similar to that of the target sample. The two techniques, as well as the Bayesian approach, were tested using 532 artificial systems in which the age marker exhibited an eight-grade expression. It was found that depending on the criteria used for evaluation, the proposed approaches and especially the one based on a hypothetical training sample, may give better results than the Bayesian method in more than 90% of the systems studied. A basic prerequisite for the good performance of the proposed approaches is to select carefully the training sample. This sample should exhibit a uniform demographic profile or a profile with almost equal numbers of young and older individuals. All the above hold if the training and the target samples have different demographic profiles. If the profiles are the same or very similar, the best aging method is the direct regression using simple linear models.

Quantitative Sex Estimation Based on Cranial Traits Using R Functions.

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.

An evaluation of Bayesian age estimation using the auricular surface in modern Greek material.

Pelvic morphology is highly reflective of both sex and age changes in humans, making it a popular research focus in forensic anthropology. Relevant studies range from traditional descriptive to more complicated approaches involving statistical modeling, with the latter having become excessively popular in the last decades. The present study examines the performance of Bayesian statistics in age estimation based on the morphological changes observed on the iliac auricular surface. The aim is two-fold: a) to test whether a Bayesian approach can improve age-at-death estimation compared to the original Lovejoy et al. (1985a) and Buckberry and Chamberlain (2002) methods, and b) to explore the impact of adopting different samples as informative priors as well as for obtaining the transition analysis parameters. For this purpose, two modern Greek documented collections have been used, the Athens and the Cretan Collection. Our results found no clear improvement in age prediction when adopting Bayesian age estimation, with only one exception: Athenian males for the Buckberry and Chamberlain (2002) method. The choice of samples for transition analysis and as informative priors affected the results but this effect was statistically non-significant.

Assessing the accuracy of cranial and pelvic ageing methods on human skeletal remains from a modern Greek assemblage.

The present study tests the accuracy of commonly adopted ageing methods based on the morphology of the pubic symphysis, auricular surface and cranial sutures. These methods are examined both in their traditional form as well as in the context of transition analysis using the ADBOU software in a modern Greek documented collection consisting of 140 individuals who lived mainly in the second half of the twentieth century and come from cemeteries in the area of Athens. The auricular surface overall produced the most accurate age estimates in our material, with different methods based on this anatomical area showing varying degrees of success for different age groups. The pubic symphysis produced accurate results primarily for young adults and the same applied to cranial sutures but the latter appeared completely inappropriate for older individuals. The use of transition analysis through the ADBOU software provided less accurate results than the corresponding traditional ageing methods in our sample. Our results are in agreement with those obtained from validation studies based on material from across the world, but certain differences identified with other studies on Greek material highlight the importance of taking into account intra- and inter-population variability in age estimation.

A quantitative approach for sex estimation based on cranial morphology.

OBJECTIVES: This paper proposes a method for the quantification of the shape of sexually dimorphic cranial traits, namely the glabella, mastoid process and external occipital protuberance. MATERIALS AND METHODS: The proposed method was developed using 165 crania from the documented Athens Collection and tested on 20 Cretan crania. It is based on digital photographs of the lateral view of the cranium, drawing of the profile of three sexually dimorphic structures and calculation of variables that express the shape of these structures. RESULTS: The combinations of variables that provide optimum discrimination between sexes are identified by means of binary logistic regression and discriminant analysis. The best cross-validated results are obtained when variables from all three structures are combined and range from 75.8 to 85.1% and 81.1 to 94.6% for males and females, respectively. The success rate is 86.3-94.1% for males and 83.9-93.5% for females when half of the sample is used for training and the rest for prediction. Correct classification for the Cretan material based upon the standards developed for the Athens sample was 80-90% for the optimum combinations of discriminant variables. DISCUSSION: The proposed method provides an effective way to capture quantitatively the shape of sexually dimorphic cranial structures; it gives more accurate results relative to other existing methods and it does not require specialized equipment. Equations for sex estimation based on combinations of variables are provided, along with instructions on how to use the method and Excel macros for calculation of discriminant variables with automated implementation of the optimum equations.

The auricular surface as age indicator in a modern Greek sample: A test of two qualitative methods.

The auricular surface is often found very well preserved, thus age-related changes in this anatomical area can be important for any set of human remains that require identification under different taphonomic conditions. This study tests the Buckberry and Chamberlain (2002) and Schmitt et al. (2005) methods in predicting the age of individuals in a documented sample from Crete, Greece. Both methods were used to record changes on the auricular surface in a mixed-sex sample of 74 individuals, directly as well as through photographs, by two independent observers. Cohen's kappa and intra class correlation coefficients (ICC) were used in order to assess inter-observer and intra-observer agreement. Results showed than none of the methods predicted age with sufficient accuracy, as high error rates were recorded. The Schmitt et al. (2005) method performed better, mainly because the age ranges it uses are broader. Scoring through photographs does not seem to introduce bias in predicting age, as demonstrated by the high intra-observer agreement rates. Inter-observer agreement was also high. The low intra- and inter-observer error rates suggest that the poor performance of both methods in the Cretan sample is not due to a lack of clarity in the description of the morphological changes recorded on the auricular surface; rather it should be attributed to a poor correlation between these changes and age at death in our material.

Population-specificity of sexual dimorphism in cranial and pelvic traits: evaluation of existing and proposal of new functions for sex assessment in a Greek assemblage.

The current paper examines the accuracy of existing binary logistic regression equations for sex prediction based on pelvic and cranial traits in a modern Greek assemblage and proposes new equations with the aim of improving correct classification rates for Balkan material. Our results suggest that existing equations based on pelvic traits perform very well on the Greek material, which can be attributed to the fact that sexual dimorphism in the pelvis results from common evolutionary forces across populations. In contrast, equations based on cranial traits are highly dependent upon the populations based on which they were developed, stressing the need to produce more population-specific functions. Our proposed equations achieve correct sex classification in 92.59% of the females and 95.79% of the males for pelvic traits, while these percentages rise to 97.53% for females and 98.95% for males when the vertical femoral head diameter is included in the models. Our functions based on cranial traits produced correct classifications in up to 92.59% of females and 88.42% of males, and when the cranial scores where combined with the vertical femoral head diameter, the correct classification rates increased to 93.83% for females and 94.73% for males. Prior to the generalization of the use of these functions, further research examining their accuracy in other groups is required, but our results appear promising.

Beyond food: The multiple pathways for inclusion of materials into ancient dental calculus.

Dental calculus (mineralized dental plaque) was first recognised as a potentially useful archaeological deposit in the 1970s, though interest in human dental calculus as a resource material has increased sharply in the past few years. The majority of recent research has focused on the retrieval of plant microfossils embedded in its matrix and interpretation of these finds as largely the result of deliberate consumption of plant-derived food. However, while most of the material described in published works does represent food, dental calculus is in fact a "depositional environment" as material can enter the mouth from a range of sources. In this respect, it therefore represents an archaeological deposit that can also contain extensive non-dietary debris. This can comprise a wide variety of cultural and environmental material which reaches the mouth and can become embedded in dental calculus through alternative pathways. Here, we explore the human behaviors and activities besides eating that can generate a flux of particles into the human mouth, the broad range of additional cultural and environmental information that can be obtained through the analysis and contextualisation of this material, and the implications of the additional pathways by which material can become embedded in dental calculus.

An attempt to predict tooth crown height using cementoenamel junction diameters and implications in the quantitative study of dental wear.

SUMMARY: Dental wear has been used extensively in osteoarchaeological studies as a means of exploring dietary patterns and the use of teeth as tools. The present paper aims to test whether crown height can be predicted for premolars and molars based on the mesiodistal and buccolingual diameters of these teeth at the level of the cementoenamel junction (CEJ). Multiple linear regression was used with crown height as dependent variable and diameters at the CEJ as predictors. Our results found that there are cases where the impact of one predictor appears to be statistically significant. However, even in these cases the 95% confidence interval for the predicted value is so broad that makes the accurate determination of the crown height based on the buccolingual or/and mesiodistal diameters at the CEJ impossible.