BRAR-anchored multimodal dataset of panoramic radiographs for periodontal bone resorption grading.

Accurate assessment of alveolar bone resorption from panoramic radiographs is crucial for diagnosing and managing periodontal disease. However, the development of robust, automated diagnostic tools is hindered by a lack of large-scale, expertly annotated multimodal datasets. To address this, we present a comprehensive dataset from 1,104 patients, integrating demographic information, full-mouth panoramic radiographs, and clinician-annotated clinical indicators at both patient and tooth levels. A key feature is the Bone Resorption Age Ratio (BRAR), which standardizes bone resorption assessment. Annotations were validated by experienced periodontists to ensure high data quality. This dataset, with its rich multimodal structure and expert validation, provides a foundational resource to accelerate AI-driven diagnostic model development and large-scale clinical research in periodontics, advancing periodontal disease diagnosis, prognosis, and epidemiology.