Automating Postmortem Interval Estimation with Computational Forensic Tools.

Abstract

JOSH GEORGE, College of Science and Technology, Fairmont State University, Fairmont, WV, 26554, KYLER SAUNDERS, College of Business and Aviation , Fairmont State University, Fairmont, WV, 26554, CHAD TIFFNER, College of Science and Technology, Fairmont State University, Fairmont, WV, 26554, and CHLOE MITCHELL, College of Science and Technology , Fairmont State University, Fairmont, WV, 26554, KRISTY HENSON, College of Science and Technology, Fairmont State University, Fairmont, WV, 26554,

Automating Postmortem Interval Estimation with Computational Forensic Tools.

Postmortem interval (PMI) estimation plays a crucial role in forensic death investigations, traditionally relying on mathematical models such as the Hennessy nomogram, the Glaister Equation, and morphological taphonomic changes. However, these methods are labor-intensive, prone to human error, and lack automation. This project presents a computational approach to PMI estimation, integrating existing mathematical models and morphological changes used by state death investigators with machine learning algorithms. Currently we are developing a user-friendly web application to implement this technology. This system will allow death investigators to incorporate morphological taphonomic data with abiotic variables while automatically adjusting PMI estimates through predictive modeling resulting in a time of death estimate based on date found and date last seen alive. FastAPI for backend processing and Angular for frontend interface are being used. Prototype results are undergoing validation with historical forensic cases and synthetic datasets, demonstrating significant reductions in human error and processing time compared to manual calculations.

This research was made possible by NASA West Virginia Space Grant Consortium, Training Grant #80NSSC20M0055. .