Estimating Age and Gender Using Artificial Intelligence in Digital Forensics

Abstract

Accurate identification of human remains is critical in forensic science, yet traditional morphological methods are subjective and limited. Post-mortem computed tomography (PMCT) offers non-invasive, high-resolution skeletal analysis, but requires advanced tools for data interpretation. This study explores machine learning (ML) to objectively analyze PMCT scans for estimating age and sex, enhancing reliability in biological profiling. We evaluate the diagnostic values of skeletal features, compare anatomical regions (skull, torso, limbs), and validate ML models for forensic applications. By integrating computational methods with PMCT, the study aims to establish standardized, data-driven techniques, advancing accuracy and reproducibility in forensic anthropology for criminal, disaster, and archaeological investigations.