Abstract:When a disaster occurs, the authority must prioritize two things. First, the search and rescue of lives, and second, the identification and management of deceased individuals. However, with thousands of dead bodies to be individually identified in mass disasters, forensic teams face challenges such as long working hours resulting in a delayed identification process and a public health concern caused by the decomposition of the body. Traditional manual dental age estimation methods are time-consuming, especially when dealing with a large number of victims. The study proposes the use of artificial intelligence (AI) to automate this process, introducing the Forensic Dental Estimation Lab (F-DentEst Lab), which employs deep convolutional neural networks to estimate dental age from digital panoramic images. The study aims to test the model's performance on Malaysian children based on a large, out-of-sample dataset (n=4892). F-DentEst Lab significantly improves efficiency, with dental age estimation taking less than 10 seconds per sample. The system features a user-friendly interface with customizable parameters. One-thousand-four-hundred digital dental panoramic images were used for training and testing, with an 80% and 20% allocations, respectively. Overall, F-DentEst Lab presents a promising AI-driven solution to enhance the efficiency of forensic dental age estimation in mass disaster scenarios.

Development and validation of automated Forensic Dental Age Estimation Lab (F-DentEst Lab)