Summary of the Need for the Project

Periapical lesions are among the most common maxillofacial and dental pathologies. If not diagnosed and treated in a timely manner, they may lead to alveolar bone destruction and eventual tooth loss. Cone-Beam Computed Tomography (CBCT) serves as a valuable three-dimensional diagnostic tool, providing detailed visualization of maxillofacial structures. However, manual interpretation of CBCT images is often time-consuming and challenging due to the large volume of data, the complexity of anatomical structures, and the potential for human error. In recent years, the application of artificial intelligence, particularly deep learning algorithms, in medical image analysis has significantly improved diagnostic accuracy and efficiency. Among these algorithms, YOLO (You Only Look Once) has emerged as a promising approach because of its advanced architecture and its ability to rapidly and accurately detect visual patterns. Consequently, YOLO represents a suitable candidate for the automated identification of periapical lesions. Despite the growing interest in AI-based diagnostic systems, only a limited number of studies have evaluated the performance of YOLO in comparison with well-established deep learning models such as ResNet for the analysis of three-dimensional CBCT data. Therefore, conducting this research is essential to develop an accurate, efficient, and reliable framework for the automated detection of periapical lesions. The findings of this study may contribute to improving diagnostic quality, supporting clinical decision-making, and advancing the integration of artificial intelligence into dental practice.

Principal Investigator: Dr. Farzaneh Mosaavat

Project Start Year: 2026