A review of manual, semi-automatic, and automated methods in dental radiographic analysis for clinical and forensic diagnosis

Abstract

Dental radiography serves as a crucial tool in clinical diagnosis, treatment planning, and forensic identification. Advances in imaging technologies such as panoramic radiography orthopantomogram (OPG), periapical imaging, and cone-beam computed tomography (CBCT) have significantly enhanced diagnostic capabilities. In artificial intelligence (AI) and deep learning (DL) methods have transformed dental image analysis, enabling automated feature extraction, segmentation, classification, and quality assessment with promising clinical reliability. This review systematically synthesizes studies published between 2020 and 2025 across Google Scholar, PubMed, ScienceDirect, Web of Science, and IEEE Xplore. The literature is categorized into six themes: i) imaging modalities and clinical applications, ii) morphometric and forensic analyses, iii) dental age estimation, iv) anatomical landmark detection, v) AI-based diagnostic systems, and vi) image quality evaluation. Findings reveal that convolutional neural networks (CNNs), transformer-based architectures, and hybrid models have achieved high diagnostic accuracy, often surpassing manual assessments. However, challenges remain regarding dataset diversity, acquisition variability, and cross-population generalizability. Addressing these limitations requires the development of standardized, multi-center datasets and domain-adaptive learning frameworks to improve robustness and clinical translation. Overall, this review highlights current progress and research gaps in AI dental radiographic analysis and emphasizes the importance of integrating data standardization and domain adaptation within intelligent dental imaging.