U.S. flag

An official website of the United States government, Department of Justice.

Developing, Visualization, and Reporting a Combined 2D+3D Automated Dental Biometrics Identification System for Forensic Odontology

Award Information

Award #
Funding Category
Competitive Discretionary
Congressional District
Funding First Awarded
Total funding (to date)

Description of original award (Fiscal Year 2019, $100,000)

Law enforcement agencies have been developing dental biometric feature-based tools in forensic identification, including identifying the victims of massive catastrophes, such as the 9/11 terrorist attack, airplane crashes, and other natural disasters such as hurricanes and tsunamis. When human remains are found, the priority of the investigation is to confirm the identity of the deceased. Physical features are often destroyed in such situations, making it very difficult to identify an individual. Dental tissues have the capability of withstanding higher temperatures and pressure. The survivability of dental features makes them an ideal identification tool. Forensic dental identification methods typically use radiological documentation techniques such as periapical radiographs, bitewing films, panoramic radiographs, and computed tomography (CT). Forensic dentists routinely use 2-dimensional (2D) static imaging techniques to generate a mental picture of the anatomy. Though viable, the deepness of internal structure cannot be localized. Moreover, the quantity and quality of antemortem dental records is extremely inconsistent across the world. Hence, analyzing and visualizing different scans is tedious and time consuming. In some cases, features extracted from dental x-rays are insufficient to identify individuals. Also, different imaging techniques carry varying amounts of information. For example, bitewing images hold localized information about the curvature of the tooth and the roots, whereas, panoramic images illustrate an end to end picture. The researcher proposes a solution that will develop Artificial Intelligence (AI) based system capable of utilizing different sensor modalities to create a robust identification system (2D and 3-dimensional (3D) model) to aid in human identification. The methods will focus on visible and radiographic 2D and 3D modeling, analysis, and visualization.

Note: This project contains a research and/or development component, as defined in applicable law," and complies with Part 200 Uniform Requirements - 2 CFR 200.210(a)(14). ca/ncf

Date Created: September 16, 2019