Related Experiment Video
Updated: Oct 24, 2026

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
Technical note: Introducing FoRDent, a new tool for estimating population affinity from dental morphology
Heather J H Edgar1, Joseph T Hefner2
1Department of Anthropology and Office of the Medical Investigator, University of New Mexico, MSC01 1040, Albuquerque, NM, 87131, USA.
Abstract:
Along with age, sex, and stature, population affinity is a core component of the biological profile anthropologists estimate for unidentified human remains. FoRDent is a new tool for estimating population affinity from dental morphology. FoRDent provides: 1) a simple interface for practitioners with varying levels of experience with dental morphological data; 2) population affinity estimations reflecting contemporary thinking in biological anthropology; 3) results from machine learning well-suited to the categorical nature of the complex, multidimensional nature of dental morphological characteristics. We describe how FoRDent meets these goals and provides a user-friendly, robust new tool for practitioners of forensic anthropology, especially in the United States and Mexico. Drawing on a large, diverse sample (n = 6052) users score up to 65 variables. To reflect the global-scale, national, and regional nature of concepts of population variation, FoRDent returns affinity results at three levels, Continental, US, and Finer resolution/20th-21st century. Accuracy decreases with the specificity of the model used due to the number of potential classifications and the fuzzy nature of population affinity classification, especially at local levels. Balanced accuracy for the three models was 64% for the four groups in the Continental model, 58% for the six groups in the US-focused model, and 26% for the 23 groups in the Finer resolution model. FoRDent provides a user-friendly platform integrating images for trait scoring, case-specific modeling, transparent diagnostics, and multi-level reporting for population affinity estimation.

