Applications, capabilities, and limitations of artificial intelligence in forensic odontology: a scoping review protocol

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Introduction: Forensic odontology plays an essential role in legal investigations by facilitating human identification through dental records, bite marks, and age estimation. Historically dependent on expert-driven, manual evaluations, this field is being reshaped by technological advancements. The integration of artificial intelligence, including machine learning, deep learning, and computer vision—offers new possibilities to automate and enhance analyses, improving diagnostic accuracy, accelerating case processing, and reducing subjectivity. Objectives: The general objective of this study is to map and understand the implications, capabilities, and limitations of artificial intelligence technologies for forensic practitioners and legal stakeholders. To facilitate a targeted data extraction strategy, this review will address specific operational objectives, namely: to identify the primary areas of forensic dentistry where artificial intelligence is currently applied; to assess the reported capabilities and performance outcomes of various artificial intelligence models in these applications, to categorize the technical limitations and challenges reported in the literature, such as data quality, sample size, and lack of standardization, and to map the ethical and legal implications surrounding the use of AI in forensic contexts, including algorithm transparency and biometric data privacy. Methodology: A scoping review will be performed according to the Joanna Briggs Institute methodology and PRISMA-ScR guidelines. The protocol is registered on the Open Science Framework. Searches will be conducted across five scientific databases and grey literature, using specific keywords. Studies published between 2023 and 2025 in English, Portuguese, Spanish, and French will be considered. Study selection and screening will be managed via Rayyan. Results: Expect to find multiple artificial intelligence techniques applied to forensic odontology, including machine learning and deep learning models for age and sex estimation. The review will likely demonstrate that AI can significantly enhance diagnostic accuracy and analysis speed, although its performance may be limited by dataset heterogeneity and a lack of standardization. Conclusion: This scoping review will clarify the current state of evidence regarding the use of artificial intelligence in forensic identification. The findings may guide future research and support the development of ethical, standardized frameworks for implementing these technologies in medico-legal investigations.