SPINNE: An app for human vertebral height estimation based on artificial neural networks.

The absence or poor preservation of vertebrae often prevent the application of the anatomical method for stature estimation. The main objective of this paper was to develop a web app based on artificial neural network (ANN) models to estimate the vertebral height of absent or poorly preserved vertebrae from other vertebrae and thus enable the application of anatomical methods. Artificial neural models were developed based on the vertebral height of vertebrae C2 to S1 of a sample composed of 56 adult male and 69 adult female individuals. The skeletons belong to the Identified Skeletal Collection of the University of Coimbra and the ages at death of these individuals ranged from 22 to 58 years old. Statistical analysis and algorithmic development were performed with the R language, R Core Team (2018). Intra- and inter-observer errors regarding the vertebral height were small for all vertebrae (<0.45 mm). Significant models to estimate vertebral height were obtained for both sexes and for the sex-pooled group, although none with an R2 higher than 0.48 and 0.34 for the C2 and the S1, respectively. The root mean square error (RMSE) regarding the predicted vertebral height and the observed vertebral height was almost always smaller than 1.0 mm while most R2 values were higher than 0.6 although models with worse performances were obtained for some vertebrae located at the ends of the vertebral column (C3, L4, and L5). The ANN models have clear potential to predict vertebral height. This mathematical approach may be used to enable the application of the anatomical method for stature estimation when some vertebrae are absent or poorly preserved. The application of the ANN models can be carried out by using the new web based app SPINNE.

Dead weight: Validation of mass regression equations on experimentally burned skeletal remains to assess skeleton completeness.

In very fragmentary remains, the thorough inventory of skeletal elements is often impossible to accomplish. Mass has been used instead to assess the completeness of the skeleton. Two different mass-based methods of assessing skeleton completeness were tested on a sample of experimentally burned skeletons with the objective of determining which of them is more reliable. The first method was based on a simple comparison of the mass of each individual skeleton with previously published mass references. The second method was based on mass linear regressions from individual bones to estimate complete skeleton mass. The clavicle, humerus, femur, patella, metacarpal, metatarsal and tarsal bones were used. The sample was composed of 20 experimentally burned skeletons from 10 males and 10 females with ages-at-death between 68 and 90years old. Results demonstrated that the regression approach is more objective and more reliable than the reference comparison approach even though not all bones provided satisfactory estimations of the complete skeleton mass. The femur, humerus and patella provided the best performances among the individual bones. The estimations based on the latter had root mean squared errors (RMSE) smaller than 300g. Results demonstrated that the regression approach is quite promising although the patella was the only reasonable predictor expected to survive sufficiently intact to a burning event at high temperatures. The mass comparison approach has the advantage of not depending on the preservation of individual bones. Whenever bones are intact though, the application of mass regressions should be preferentially used because it is less subjective.

MEPROCS framework for Craniofacial Superimposition: Validation study.

Craniofacial Superimposition (CFS) involves the process of overlaying a skull with a number of ante-mortem images of an individual and the analysis of their morphological correspondence. The lack of unified working protocols and the absence of commonly accepted standards, led to contradictory consensus regarding its reliability. One of the more important aims of 'New Methodologies and Protocols of Forensic Identification by Craniofacial Superimposition (MEPROCS)' project was to propose a common framework for CFS, what can be considered the first international standard in the field. The framework aimed to serve as a roadmap for avoiding particular assumptions that could bias the process. At the same time, it provides some empirical support to certain practices, technological means, and morphological criteria expected to facilitate the application of the CFS task and to improve its reliability. In order to confirm the utility and potential benefits of the framework use, there is a need to empirically evaluate it in CFS identification scenarios as close as possible to the reality. Thus, the purpose of this study is to validate the CFS framework developed. For that aim 12 participants were asked to report about a variable number of CFS following all the recommendations of the framework. The results are analysed and discussed according to the framework understanding and fulfilment, the participants' performance, and the correlation between expected decisions and those given by the participants. In view of the quantitative results and qualitative examination criteria we can conclude that those who follow the MEPROCS recommendations improve their performance.

Study on the performance of different craniofacial superimposition approaches (I).

As part of the scientific tasks coordinated throughout The 'New Methodologies and Protocols of Forensic Identification by Craniofacial Superimposition (MEPROCS)' project, the current study aims to analyse the performance of a diverse set of CFS methodologies and the corresponding technical approaches when dealing with a common dataset of real-world cases. Thus, a multiple-lab study on craniofacial superimposition has been carried out for the first time. In particular, 26 participants from 17 different institutions in 13 countries were asked to deal with 14 identification scenarios, some of them involving the comparison of multiple candidates and unknown skulls. In total, 60 craniofacial superimposition problems divided in two set of females and males. Each participant follow her/his own methodology and employed her/his particular technological means. For each single case they were asked to report the final identification decision (either positive or negative) along with the rationale supporting the decision and at least one image illustrating the overlay/superimposition outcome. This study is expected to provide important insights to better understand the most convenient characteristics of every method included in this study.