Integrating artificial intelligence and adult dental age estimation in forensic identification: A literature review

Abstract

Age estimation is crucial in various forensic fields, including forensic medicine, anthropology, and demographic studies. Adult dental age estimation is affected by multiple factors, resulting in discrepancies between dental age and chronological age. The development of artificial intelligence (AI) technology has led to extensive investigations in forensic sciences, encompassing several areas such as facial recognition, age, sex identification, and DNA analysis. Adult age estimation methods commonly used include the pulp-tooth ratio approach, the Harris & Nortje method, and the Van Heerden method. AI approaches such as Fuzzy Logic (FL), Evolutionary Computing (EC), and Machine Learning (ML) are being extensively applied. These techniques use algorithms to imitate human thinking and behavior. Deep learning techniques, explicitly using deep convolutional neural networks (DCNN), enable age estimation by segmenting images and making measurements, replicating the cognitive processes of radiologists when computing indices such as the third molar maturity (I3M) index. Also, DCNNs automatically optimize teeth segmentation in dental X-ray images, improving image refining and analysis efficiency. AI integration in forensic dentistry improves the precision and effectiveness of dental data processing while significantly accelerating individual identification procedures. Incorporating this technology shows potential for enhancing the caliber and dependability of evidence in forensic investigations.