Annotated Dental Panoramic Radiograph Dataset
Year: 2026
Type: dataset
Abstract: This dataset contains 1,667 annotated panoramic dental radiographs collected from a clinical dental setting and used in the study "Comparative Analysis of Object Detection Models on Panoramic Radiographs." Images were acquired from patients attending Başkent University Faculty of Dentistry and anonymized prior to release; all patient identifiers, acquisition dates, and identifying file names were removed. The study was approved by the Başkent University Non-Interventional Clinical Research Ethics Committee (Project No: KA25/460, Decision No: 26/08, dated 14 January 2026).The dataset contains 21,187 bounding-box annotations across six clinically relevant classes: Crown, Filling, Root Canal Treatment, Implant, Third Molar Tooth, and Caries. Annotations were performed on the Roboflow platform by an informatics specialist under the supervision of two experienced dentomaxillofacial radiologists and one dentist. The dataset is provided in two annotation formats: COCO JSON (used for Faster R-CNN training) and YOLOv11 (used for RT-DETR-L, YOLO26-L, and YOLOv12-X training). Images are split into a training/validation pool (85%, n = 1,417) and a held-out test set (15%, n = 250).
Source: Figshare
Language: English
Subfield: Oral Surgery
Field: Dentistry
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7166881311