CTooth+: A Large-Scale Dental Cone Beam Computed Tomography Dataset and Benchmark for Tooth Volume Segmentation

Accurate tooth volume segmentation is a prerequisite for computer-aided dental analysis. Deep learning-based methods have achieved satisfying performances but require large quantity of data with ground truth. The publicly available limited meaning the existing can not be reproduced, evaluated and applied in clinical practice. In this paper, we establish 3D CBCT dataset CTooth+, 22 fully annotated volumes 146 unlabeled volumes. We further evaluate several state-of-the-art strategies based on fully-supervised learning, semi-supervised learning active define performance principles. This work provides new benchmark task, experiment serve as baseline future AI-based imaging research application development. codebase are released here.