Adults’ dental cone beam computed tomography images dataset for detecting and classifying missing teeth

The zip file named "Dataset" contains our annotated dataset, which includes annotated missing teeth sites from 85 patients with 30,423 Cone Beam Computed Tomography (CBCT) slices, along with unannotated missing teeth sites from 73 patients with 29,555 CBCT slices. This resource is the first to provide three-dimensional (3D) annotations of missing teeth on CBCT images, enabling the training of deep learning models for automated implant planning. The zip file named "Code" contains the code used in the technical validation section (submit_paper) and a lightweight Python script (dicom_to_png_converter.py). This script automatically traverses the dataset folders, reads 3D DICOM series, and converts all slices sequentially into standard 2D image files (e.g., PNG format) with proper global contrast normalization. The zip file named "Visualization cases" contains several representative cases from different vendors (e.g., KaVo), and we uploaded their fully converted 2D PNG slices for immediate visual inspection.