Caries-Spectra: A dataset of Enamel Caries

Published: 18 December 2023| Version 2 | DOI: 10.17632/9jnf2jvghy.2


This dataset comprises a diverse collection of low-resolution images featuring two types of dental caries along with a category that represents no caries. Carefully selected from different patients, these caries offer a comprehensive and varied representation. The images were sourced from several medical clinics from Rajshahi and Dhaka, Bangladesh, spanning the time period from 2021 to 2022. Functioning as a valuable visual resource, the dataset provides a detailed portrayal of the distinctive characteristics of these caries, facilitating accurate classification. It is divided into two specific categories: Advanced Enamel Caries (800 images) and Early Stage Enamel Caries (800 images), and No Enamel Caries (400 images) totaling 2000 original JPG images. The dataset is organized into three sub-folders: "AdvanceEnamel_Caries" and "EarlyStageEnamel_Caries, and "NoEnamel_Caries".



American International University Bangladesh, Bangladesh University of Business and Technology


Computer Vision, Orthodontics, Image Processing, Machine Learning, Biomedical Imaging, Dental Caries