Quantifying the Impact of Intrinsic and Extrinsic Factors to Improve Juvenile Age Estimation

As submitted by the proposer:

Well-documented and sequential growth and development markers are used as age indicators to estimate juvenile age in forensic contexts. Many factors, referred to as covariates, are known to influence growth and development, but the interactive magnitude of their impact on age estimation is not documented. The objectives of this research are to evaluate the impact of intrinsic factors (sex, population affiliation, and indicators of pathology/stress) and extrinsic factors of variation (Human Development Index [HDI] and Gini index, as indicators of life quality and social inequality, respectively) on juvenile age estimation at both local and global levels.

Data will be collected from numerous socially and geographically diverse populations including Angola, Colombia, Denmark, the United States, France, the Netherlands, South Africa, and Singapore. The skeletal (diaphyseal and pelvic dimensions, epiphyseal fusion) and dental development data will be collected primarily from computed tomography (CT) images from hospitals and medical examiner’s offices, but also from skeletal collections, of individuals aged 0 to 15 years (n >3,500 individuals).

Transition analysis and neural network algorithms will be used to construct population-specific and universal models for age estimation. A hierarchical linear model will be used to quantify the impact of the country-level HDI and Gini index on each covariate and age indicator to evaluate and interpret the effects of a country’s SES.

This project will vertically advance juvenile age estimation by providing answers to questions concerning the impact of covariates on age estimation. This will provide strong evidence for or against the use of specific or global methods in forensic anthropology. Data will be recorded and distributed in previously developed graphical user interfaces, KidStats (NIJ Award 2015-DN-BX-K009) and KSCollect (NSF BCS-1551913), both of which are freely available. Improvements to KidStats will include incorporation of the additional six reference samples and allow the practitioner to build population specific age estimation models, geographic or SES-specific models, or combine all reference samples to build a global model. This will ensure applicability around the world, and have great impact in the scientific and human rights communities.

This timely and innovative project addresses contemporary social issues resulting from major human movement events as well as response capabilities for natural and mass disasters. The project, and specifically the data, will have far-reaching implications that will foster and facilitate multidisciplinary, national, and international collaborations regarding a multitude of research questions independent of the specific aims of the current research.

Note: This project contains a research and/or development component, as defined in applicable law.

ca/ncf