A novel framework for integrating post-mortem images with 3D body models for automatic injury medical documentation.

Forensic medicine is critical in criminal investigations, providing essential insights into the nature and mechanisms of injuries. Traditional post-mortem examination remains mostly manual, which introduces risks of error and inefficiency, particularly in injury assessment. This study proposes a fully automated system for detecting and mapping injuries (bruises and abrasions) on a 3D human body model to improve forensic documentation. The developed automated system consisted of the following modules: (1) a module for predicting injuries in images, (2) a module for generating a 3D human body based on photos, (3) a module for integrating detected injuries from the images with the 3D model, and (4) a module for performing measurements on the integrated 3D human model with the detected injuries. It enhances the objectivity of injury documentation, reduces human error, preserves digital evidence, and demonstrates a strong potential for integration into medical technologies that support forensic pathology and postmortem diagnostics.