Body mass estimation in modern population using anthropometric measurements from computed tomography

Technical note: Prediction of body mass from stature and pelvic breadth

Equations for predicting body mass from stature and bi-iliac (maximum pelvic) breadth have been developed, but have had variable success when applied to living or recently deceased individuals, calling into question their general applicability. Here we test these equations on a large, ethnically diverse sample. Skeletal and anthropometric data for 507 recently deceased Indigenous, Hispanic, and non-Hispanic White adults were obtained from the New Mexico Decedent Image Database. The body mass of individuals with a "normal" body mass index (BMI = 18.5-24.9) is very accurately predicted, with an average directional bias of about 1% and an average random error of less than 8%. Underweight individuals (BMI < 18.5) are overpredicted, while overweight (BMI = 25-29.9) and especially obese (BMI≥30) individuals are underpredicted. Within BMI categories, there is a strong and isometric relationship between predicted and true body mass. Individual body mass prediction errors using the stature/bi-iliac method are mainly dependent on variation in BMI. Because earlier humans were more likely to fall within or close to the normal BMI range, the equations should be applicable, on an individual basis, in archeological and paleontological contexts. Because of the prevalence of obesity in many modern populations, these equations are not applicable in a general forensic context. We derive new equations from nonobese individuals in our sample (n = 338), which produce reasonable average prediction errors. If obese individuals can be identified using other skeletal parameters, these equations may be useful in estimating body mass in nonobese forensic cases.

The estimation and evolution of hominin body mass

Body mass is a critical variable in many hominin evolutionary studies, with implications for reconstructing relative brain size, diet, locomotion, subsistence strategy, and social organization. We review methods that have been proposed for estimating body mass from true and trace fossils, consider their applicability in different contexts, and the appropriateness of different modern reference samples. Recently developed techniques based on a wider range of modern populations hold promise for providing more accurate estimates in earlier hominins, although uncertainties remain, particularly in non-Homo taxa. When these methods are applied to almost 300 Late Miocene through Late Pleistocene specimens, the resulting body mass estimates fall within a 25-60 kg range for early non-Homo taxa, increase in early Homo to about 50-90 kg, then remain constant until the Terminal Pleistocene, when they decline.

Effects of obesity on talar micro- and macro-morphology

The addition of information regarding obesity status to the forensic anthropological biological profile could significantly contribute to the identification of human skeletal remains since over 40% of the U.S. adult population is currently obese. This study examines the differences in talar shape and trabecular bone structure between obese and non-obese individuals. A sample of 20 obese and 20 non-obese divided evenly by sex was selected from the Texas State University Donated Skeletal Collection. Tali were imaged using x-ray computed tomography (voxel size: 28-38.7 μm). Image stacks were processed to produce binary images as well as trabecular thickness and spacing maps. Landmark-based geometric morphometric analyses were conducted to quantify shape variation. Shape coordinates were used to locate 100 geometrically homologous volumes of interest within each talus. Bone volume fraction, trabecular thickness, and trabecular spacing were extracted at each volume of interest. Within each sex, a one-way ANCOVA was used to determine if significant differences exist between obese and non-obese individuals in trabecular bone after controlling for age. The size of the talus as well as subtle aspects of shape were found to distinguish the sexes. The results further indicate that bone volume fraction significantly differs between obese and non-obese males. In females, bone volume fraction is correlated with age but does not differ between obese and non-obese. The study demonstrates that bone microstructure is a promising approach to estimating body mass or body mass index category but age effects diminish the potential for the talus to be used alone.

Stature estimation by semi-automatic measurements of 3D CT images of the femur

Stature estimation is one of the most basic and important methods of personal identification. The long bones of the limbs provide the most accurate stature estimation, with the femur being one of the most useful. In all the previously reported methods of stature estimation using computed tomography (CT) images of the femur, laborious manual measurement was necessary. A semi-automatic bone measuring method can simplify this process, so we firstly reported a stature estimation process using semi-automatic bone measurement software equipped with artificial intelligence. Multiple measurements of femurs of adult Japanese cadavers were performed using automatic three-dimensional reconstructed CT images of femurs. After manually setting four points on the femur, an automatic measurement was acquired. The relationships between stature and five femoral measurements, with acceptable intraobserver and interobserver errors, were analyzed with single regression analysis using the standard error of the estimate (SEE) and the coefficient of determination (R²). The maximum length of the femur (MLF) provided the lowest SEE and the highest R²; the SEE and R² in all cadavers, males and females, respectively, were 3.913 cm (R² = 0.842), 3.664 cm (R² = 0.705), and 3.456 cm (R² = 0.686) for MLF on the right femur, and 3.837 cm (R² = 0.848), 3.667 cm (R² = 0.705), and 3.384 cm (R² = 0.699) for MLF on the left femur. These results were non-inferior to those of previous reports regarding stature estimation using the MLF. Stature estimation with this simple and time-saving method would be useful in forensic medical practice.

Belly fat or bloating? New insights into the physical appearance of St Anthony of Padua

Over the centuries, iconographic representations of St Anthony of Padua, one of the most revered saints in the Catholic world, have been inspired by literary sources, which described the Saint as either naturally corpulent or with a swollen abdomen due to dropsy (i.e. fluid accumulation in the body cavities). Even recent attempts to reconstruct the face of the Saint have yielded discordant results regarding his outward appearance. To address questions about the real appearance of St Anthony, we applied body mass estimation equations to the osteometric measurements taken in 1981, during the public recognition of the Saint's skeletal remains. Both the biomechanical and the morphometric approach were employed to solve some intrinsic limitations in the equations for body mass estimation from skeletal remains. The estimated body mass was used to assess the physique of the Saint with the body mass index. The outcomes of this investigation reveal interesting information about the body type of the Saint throughout his lifetime.

Assessment of body weight from percutaneous widths of the bones and joints-Implications in forensic and clinical examinations

BACKGROUND

Estimation of age, stature, sex, and ancestry contributes to the establishment of the biological profile of the deceased in forensic examinations. Assessment of the body weight aids in the approximation of the overall body size of the individual which may help in the forensic identification process. In clinical examinations, body weight assessment assumes importance in cases where body weight measurement is a challenging task due to illness and body deformity.

OBJECTIVE

The present research was conducted to estimate the body weight from the percutaneous width of the bones and joints with the help of prediction equations.

METHODS

The study was carried out on 344 adults (172 Females and 172 Males) aged between 18 and 25 years from the Himachal Pradesh State of North India. Eleven anthropometric measurements including height vertex, mid-arm circumference, humerus bicondylar width, transverse chest breadth, sagittal chest breadth, bi-iliac breadth, handbreadth, femur bicondylar breadth, ankle breadth, foot breadth, and body weight were taken on each individual. The sex differences were evaluated by using independent student t-test and Mann-Whitney U test and the correlation between the body weight and the anthropometric variables was investigated by using both Karl Pearson's correlation coefficient and Spearman's rank correlation coefficient depending upon the normality of the data. Regression models for the estimation of body weight were calculated. Further, a validation study was carried out to check the accuracy and utility of the derived regression models by calculating the mean absolute percent prediction error (MAPPE).

RESULTS

Significant sex differences were observed among all the anthropometric variables. The transverse chest breadth and mid-arm circumference were strongly correlated with the body weight, whereas, a good correlation was also observed in other measurements except for the ankle breadth. The SEE (Standard error of estimate) of the derived linear regression models was compared, and it was found that multiple linear regression models show better accuracy than simple linear regression models. The MAPPE was found to be less in the case of multiple linear regression models than the linear ones.

CONCLUSION

The present investigation concludes that regression models can be used in the estimation of body weight from the percutaneous measurements and joint widths with reasonable accuracy in an Indian population.

Allometric scaling and growth: Evaluation and applications in subadult body mass estimation

OBJECTIVES

Previously developed methods in subadult body mass estimation have not been tested in populations other than European-American or African-American. This study uses a contemporary Taiwanese sample to test these methods. Through evaluating their accuracy and bias, we addressed whether the allometric relationships between body mass and skeletal traits commonly used in subadult body mass estimation are conserved among different populations.

MATERIALS AND METHODS

Computed tomography scans of lower limbs from individuals aged 0-17 years old of both sexes were collected from National Taiwan University Hospital along with documented body weight. Polar second moment of area, distal femoral metaphyseal breadth, and maximum superior/inferior femoral head diameter were collected either directly from the scans or from reconstructed 3D models. Estimated body mass was compared with documented body mass to assess the performance of the equations.

RESULTS

Current methods provided good body mass estimates in Taiwanese individuals, with accuracy and bias similar to those reported in other validation studies. A tendency for increasing error with increasing age was observed for all methods. Reduced major axis regression showed the allometric relationships between different skeletal traits and body mass across different age categories can all be summarized using a common fitted line. A revised, maximum likelihood-based approach was proposed for all skeletal traits.

DISCUSSION

The results suggested that the allometric relationships between body mass and different skeletal traits are largely conserved among populations. The revised method provided improved applicability with strong underlying theoretical justifications, and potential for future improvements.

The relationship between adolescent obesity and pelvis dimensions in adulthood: a retrospective longitudinal study

Background

The effect of fat tissue on a developing individual is fundamentally different from the effect on an adult. Several changes caused by obesity during sexual maturation have an irreversible and severe negative effect (lower fertility, reduced final height, type 2 diabetes mellitus) even for those who have subsequently lost weight. Our study was focused on monitoring the skeletal structure substantially influenced by sex hormones—the pelvis. The adult pelvis is strongly sexually dimorphic, which is not the case for the juvenile pelvis; skeletal differences between sexes are not so prominent and start to manifest with the onset of puberty. Evidence from animal models and case studies of treatment of gender dysphoria suggests that estrogens have a stimulatory effect on the growth plates present on the pelvis, leading to morphological change. Male obesity, especially in puberty, is connected with hypogonadism, manifesting in low levels of testosterone, and high levels of estrogens. The goal of our study was to evaluate the influence of obesity during adolescence on the morphology of the adult pelvis in the context of androgen and estrogen status.

Sample and Methods

Our sample consists of 238 individuals (144 females, 94 males) observed after an 8 year follow-up (mean age during enrollment 15.2 years, follow-up 23.3 years). Anthropometry and body composition using bioimpedance analysis (BIA) were obtained. During the follow-up, saliva samples from male participants were also collected to estimate testosterone and estradiol levels using the salivary ELISA kit (Salimetrics LLC, State College, PA, USA).

Results

The body fat (percentage of body fat estimated using BIA) was strongly positively associated with relative pelvic breadths in adulthood (males r = 0.64; females r = 0.56, both with p < 0.001). Adulthood pelvic breadth was a highly sensitive (0.81) and specific (0.74) retrospective marker of obesity during adolescence. The complex regression model (with reduction of dimensionality) including testosterone, estradiol to testosterone ratio and body fat (adolescent and adulthood) was able to describe 54.8% variability of pelvic breadth among males.

Discussion

We observed that adults with a history of obesity from adolescence tend to have a wider dimension of the bony pelvis in adulthood. Based on the parameters of the adult pelvis, the history of obesity can be determined with satisfactory sensitivity and specificity (<70%). One of the explanations for this observation can be the influence of relatively elevated estrogens levels connected with excessive adiposity leading to a wider pelvis. However, the biomechanical stress connected with elevated body mass also has to be considered, as does the influence of physical activity and gait pattern on the skeletal build.

Comparison of 14 body size variables on adult skeletal age methods: how body mass, BMI, fat and muscle influence age estimations

Background: Skeletal age estimations consistently under-age individuals with low body mass and over-age individuals with high body mass.Aim: To assess the directional biases of body size variables, including fat and muscle tissue, on skeletal age estimates.Subjects and methods: CT scans of 420 age- and size-selected cadavers were assessed using the fourth rib and pubic symphysis age estimation methods. MANCOVA was used to test the effects of stature, body mass, circumference, and fat and muscle tissue area on age estimation biases.Results: For both methods, there were biases based on body mass, BMI, circumference and fat tissue area. Individuals with the lowest measurements for these variables were under-aged while individuals with the highest measurements were over-aged. Age estimation biases were not affected by stature, muscle area, or skeletal measurements of bi-iliac breadth, femoral head diameter, or femoral length.Conclusion: Increased body mass, and specifically fat tissue, accelerates skeletal ageing. Increased muscle area generally did not show a protective effect on skeletal ageing. The accuracy of age estimations would be improved by factoring in body mass. However, femoral head diameter and bi-iliac breadth are not good proxies for body mass. Osteological markers of obesity offer a promising new approach.

Discrepancies between reported and cadaveric body size measurements associated with a modern donated skeletal collection

Body mass and stature estimation methods used in biological anthropology require materials with known body size information. There are several types of body size data that can be associated with skeletal collections. However, discussion regarding the reliability and suitability of these types of information for anthropological research is scarce. This paper focuses on differences between reported and recorded cadaver weights and heights associated with a modern donated skeletal collection, similar to these commonly used in anthropological research. In addition, the study identifies factors that may influence these discrepancies. The results show statistically significant differences between reported and cadaver body size information. Generally, reported weights, statures and body mass indices (BMI) were greater compared to the cadaver information in this sample. However, potential effects on these discrepancies varied depending on sex and information type. Age was found to influence stature discrepancy in females, and donation type had an effect on the female weight discrepancy. The results also show that body size range (weight, stature and BMI) can contribute to these discrepancies. Even though the differences between reported and cadaver data may not be significant at the population level, the individual variation can cause misclassifications of individuals depending on the data used. This study encourages researchers using modern documented collections and their body size information to openly acknowledge the types of weight and stature data used and to discuss potential problems associated with them.

Estimating body mass and composition from proximal femur dimensions using dual energy x-ray absorptiometry

Body mass prediction from the skeleton most commonly employs femoral head diameter (FHD). However, theoretical predictions and empirical data suggest the relationship between mass and FHD is strongest in young adults, that bone dimensions reflect lean mass better than body or fat mass and that other femoral measurements may be superior. Here, we generate prediction equations for body mass and its components using femoral head, neck and proximal shaft diameters and body composition data derived from dual-energy x-ray absorptiometry (DXA) scans of young adults (n = 155, 77 females and 78 males, mean age 22.7 ± 1.3 years) from the Andhra Pradesh Children and Parents Study, Hyderabad, India. Sex-specific regression of log-transformed data on femoral measurements predicted lean mass with smaller standard errors of estimate (SEEs) than body mass (12-14% and 16-17% respectively), while none of the femoral measurements were significant predictors of fat mass. Subtrochanteric mediolateral shaft diameter gave lower SEEs for lean mass in both sexes and for body mass in males than FHD, while FHD was a better predictor of body mass in women. Our results provide further evidence that lean mass is more closely related to proximal femur dimensions than body or fat mass and that proximal shaft diameter is a better predictor than FHD of lean but not always body mass. The mechanisms underlying these relationships have implications for selecting the most appropriate measurement and reference sample for estimating body or lean mass, which also depend on the question under investigation.

Relationship between body mass, lean mass, fat mass, and limb bone cross-sectional geometry: Implications for estimating body mass and physique from the skeleton

OBJECTIVES

Estimating body mass from skeletal dimensions is widely practiced, but methods for estimating its components (lean and fat mass) are poorly developed. The ability to estimate these characteristics would offer new insights into the evolution of body composition and its variation relative to past and present health. This study investigates the potential of long bone cross-sectional properties as predictors of body, lean, and fat mass.

MATERIALS AND METHODS

Humerus, femur and tibia midshaft cross-sectional properties were measured by peripheral quantitative computed tomography in sample of young adult women (n = 105) characterized by a range of activity levels. Body composition was estimated from bioimpedance analysis.

RESULTS

Lean mass correlated most strongly with both upper and lower limb bone properties (r values up to 0.74), while fat mass showed weak correlations (r ≤ 0.29). Estimation equations generated from tibial midshaft properties indicated that lean mass could be estimated relatively reliably, with some improvement using logged data and including bone length in the models (minimum standard error of estimate = 8.9%). Body mass prediction was less reliable and fat mass only poorly predicted (standard errors of estimate ≥11.9% and >33%, respectively).

DISCUSSION

Lean mass can be predicted more reliably than body mass from limb bone cross-sectional properties. The results highlight the potential for studying evolutionary trends in lean mass from skeletal remains, and have implications for understanding the relationship between bone morphology and body mass or composition.

Body mass prediction from femoral volume and sixteen other femoral variables in the elderly: BMI and adipose tissue effects

OBJECTIVES

The frequently used prediction equations of body mass do not seem appropriate for elderly individuals. Here, we establish the relationship between femoral dimensions and known body mass in elderly individuals in order to develop prediction formulas and identify the factors affecting their accuracy.

MATERIALS AND METHODS

The body mass linear least-squares regression is based on 17 femoral dimensions, including femoral volume, and 66 individuals. Body proportion and composition effects on accuracy are analyzed by means of the body mass index (BMI) and on a subset sample (n = 25), by means of the masses of adipose, bone and muscle tissues.

RESULTS

Most variables significantly reflect body mass. Among them, six dimensions (e.g., biepicondylar breadth, femoral volume, and head femoral diameter) present percent standard errors of estimate ranging from 9.5 to 11% (r = 0.72-0.81) in normal BMI samples. Correlations are clearly lower in samples with normal and abnormal BMI [r = 0.38-0.58; % of standard error of estimate (SEE) = 17.3-19.6%] and not significantly correlated in females (femoral volume) who present high proportions of abnormal BMI and adipose tissue. In the subset, femoral volume is well correlated with bone mass (r = 0.88; %SEE = 7.9%) and lean body mass (r = 0.67; %SEE = 17.2%).

DISCUSSION

Our body mass estimation equations for elderly individuals are relevant since relatively low correlations are recurrent in studies using younger individuals of known body mass. However, age, sex, lifestyle, and skeleton considerations of studied populations can provide information about the relevance of the body mass estimation, which is dependent on the BMI classification and the proportion of adipose tissue. Our general considerations can be used for studies of younger individuals.

Accuracy and Reliability of Total Body Mass Estimation Techniques from Stature and Bi-iliac Breadth in Non-Hispanic U.S. Whites from the Bass Donated Skeletal Collection

This paper tests the fidelity of a recent method that used the NHANES III dataset as a proxy to estimate total body mass from stature and bi-iliac breadth in U.S. White males and females. The bi-iliac breadths of 230 males and 152 females identified as non-Hispanic U.S. White from the Bass Donated Skeletal Collection were measured, and along with stature from predonor paperwork, total body mass estimates were calculated and then compared to body masses recorded on predonor paperwork. Male and female samples were subdivided by body mass index (BMI [kg/m² ]) categories established by the World Health Organization. Our results suggest that total body mass estimates can be accurately assessed provided that the individual is within 18.50 ≤ BMI ≤ 29.99 for White males and 18.50 ≤ BMI ≤ 24.99 for White females. Recommendations on how to report total body mass estimates are also presented.

Body mass estimation from the skeleton: An evaluation of 11 methods

Estimating an individual body mass (BM) from the skeleton is a challenge for forensic anthropology. However, identifying someone's BMI (Body Mass Index) category, i.e. underweight, normal, overweight or obese, could contribute to identification. Individual BM is also known to influence the age-at-death estimation from the skeleton. Several methods are regularly used by both archaeologists and forensic practitioners to estimate individual BM. The most commonly used methods are based on femoral head breadth, or stature and bi-iliac breadth. However, those methods have been created from mean population BMs and are therefore meant to estimate the average BM of a population. Being that they are based on individual BM data and estimated femoral cortical areas, the newest published methods are supposed to be more accurate. We evaluated the accuracy and reliability of the most commonly used and most recent BM estimation methods (n=11) on a sample of 64 individuals. Both sexes and all BMI categories are represented, as well as a wide range of BM. Ages in this sample range from 20 to 87 years of age. Absolute and real differences between actual BM and estimated BM were assessed; they determined the accuracy for individual BM estimation and for average BM estimation of a population, respectively. The proportion of the sample whose estimated BM falls within ±10% and ±20% of their actual BM determines the reliability of the methods in our sample for, respectively, individual BM estimation and average BM of a population. The tested methods result in an absolute difference of 11kg-26kg±10kg with regards to prediction of individuals actual BM. The real differences are very variable from method to method, ranging from -14kg to 25kg. None of the tested methods is able to estimate BM of half of the sample within ±10% of their actual BM but most of them can estimate BM of more than half of the sample within ±20% of their actual BM. The errors increase with increasing BM, demonstrating a bias in all the methods. No bone variable tested correlated with BM. BMI categories were correctly predicted for less than 50% of the sample in most cases. In conclusion, our study demonstrates that the 11 methods tested are not suited for estimating individual BM or for predicting BMI categories. However, they are accurate and reliable enough for estimating the average BM of a population.

The evolution of body size and shape in the human career

Body size is a fundamental biological property of organisms, and documenting body size variation in hominin evolution is an important goal of palaeoanthropology. Estimating body mass appears deceptively simple but is laden with theoretical and pragmatic assumptions about best predictors and the most appropriate reference samples. Modern human training samples with known masses are arguably the 'best' for estimating size in early bipedal hominins such as the australopiths and all members of the genus Homo, but it is not clear if they are the most appropriate priors for reconstructing the size of the earliest putative hominins such as Orrorin and Ardipithecus The trajectory of body size evolution in the early part of the human career is reviewed here and found to be complex and nonlinear. Australopith body size varies enormously across both space and time. The pre-erectus early Homo fossil record from Africa is poor and dominated by relatively small-bodied individuals, implying that the emergence of the genus Homo is probably not linked to an increase in body size or unprecedented increases in size variation. Body size differences alone cannot explain the observed variation in hominin body shape, especially when examined in the context of small fossil hominins and pygmy modern humans.This article is part of the themed issue 'Major transitions in human evolution'.

Virtual anthropology - a brief review of the literature and history of computed tomography

Computed tomography (CT) has influenced numerous fields since its inception in the 1970s. The field of palaeoanthropology significantly benefited from this efficient and non-invasive medium in terms of the conservation, reconstruction and analysis of fossil human remains. Over the past decade, there has been a steady increase in the number of forensic anthropological studies incorporating virtual osteological analyses. Because of the increasing importance of these modern cross-sectional imaging techniques and the requirement for standardized parameters in forensic science, we deemed it important to outline the history and development of CT applications in these related academic areas. The present paper outlines the history of “virtual anthropology” and osteological multi-detector CT in the context of palaeoanthropology and forensic anthropology.

Inaccuracy and bias in adult skeletal age estimation: Assessing the reliability of eight methods on individuals of varying body sizes

Accurate age estimations are essential for identifying human skeletal remains and narrowing missing persons searches. This study examines how BMI, body mass, and stature influence inaccuracy and bias in adult skeletal age estimations obtained using eight methods. 746 skeletons from the Hamann-Todd and William Bass Collections were used. Underweight BMI, light body mass, and short-stature individuals have the most error associated with their age estimates and are consistently under-aged between 3 to 13years. Obese BMI, heavy body mass, and tall-stature individuals are consistently over-aged between 3 to 8.5years. The most reliable methods for smaller-bodied individuals are Kunos et al. (first rib) and Buckberry-Chamberlain (auricular surface); for individuals in the average range, İşcan et al. (fourth ribs) and Passalacqua (sacrum); and for larger-bodied individuals, İşcan et al., Passalacqua, and Rougé-Maillart et al. (auricular surface and acetabulum). Lovejoy et al. (auricular surface) and Suchey-Brooks (pubic symphysis) produce consistent inaccuracy and bias scores across all body size groups. The least reliable method for smaller-bodied individuals is İşcan et al.; for larger-bodied individuals, Buckberry-Chamberlain; and across all body size groups, DiGangi et al. (first rib).

The accuracy of body mass prediction for elderly specimens: Implications for paleoanthropology and legal medicine

Different practices in paleoanthropology and legal medicine raise questions concerning the robustness of body mass (BM) prediction. Integrating personal identification from body mass estimation with skeleton is not a classic approach in legal medicine. The originality of our study is the use of an elderly sample in order to push prediction methods to their limits and to discuss about implications in paleoanthropology and legal medicine. The aim is to observe the accuracy of BM prediction in relation to the body mass index (BMI, index of classification) using five femoral head (FH) methods and one shaft (FSH) method. The sample is composed of 41 dry femurs obtained from dissection where age (c. 82 years) and gender are known, and weight (c. 59.5 kg) and height are measured upon admission to the body leg service. We show that the estimation of the mean BM of the elderly sample is not significantly different to the real mean BM when the appropriate formula is used for the femoral head diameter. In fact, the best prediction is obtained with the McHenry formula (1992), which was based on a sample with an equivalent average mass to that of our sample. In comparison, external shaft diameters, which are known to be more influenced by mechanical stimuli than femoral head diameters, yield less satisfactory results with the McHenry formula (1992) for shaft diameters. Based on all the methods used and the distinctive selected sample, overestimation (always observed with the different femoral head methods) can be restricted to 1.1%. The observed overestimation with the shaft method can be restricted to 7%. However, the estimation of individual BM is much less reliable. The BMI has a strong impact on the accuracy of individual BM prediction, and is unquestionably more reliable for individuals with normal BMI (9.6% vs 16.7% for the best prediction error). In this case, the FH method is also the better predictive method but not if we integrate the total sample (i.e., the FSH method is better with more varied BMI). Finally, the estimation of the mean BM of a sample can be used with more confidence compared to the estimation of individual BM. The former is very useful in an evolutionary perspective whereas the latter should be used in keeping with the information gathered on the studied specimen in order to reduce prediction errors. Finally, the BM estimation can be a parameter to consider for personal identification.

Selecting best-fit models for estimating the body mass from 3D data of the human calcaneus

Body mass (BM) estimation could facilitate the interpretation of skeletal materials in terms of the individual's body size and physique in forensic anthropology. However, few metric studies have tried to estimate BM by focusing on prominent biomechanical properties of the calcaneus. The purpose of this study was to prepare best-fit models for estimating BM from the 3D human calcaneus by two major linear regression analysis (the heuristic statistical and all-possible-regressions techniques) and validate the models through predicted residual sum of squares (PRESS) statistics. A metric analysis was conducted based on 70 human calcaneus samples (29 males and 41 females) taken from 3D models in the Digital Korean Database and 10 variables were measured for each sample. Three best-fit models were postulated by F-statistics, Mallows' Cp, and Akaike information criterion (AIC) and Bayes information criterion (BIC) for each available candidate models. Finally, the most accurate regression model yields lowest %SEE and 0.843 of R(2). Through the application of leave-one-out cross validation, the predictive power was indicated a high level of validation accuracy. This study also confirms that the equations for estimating BM using 3D models of human calcaneus will be helpful to establish identification in forensic cases with consistent reliability.