Abstract

Accurate delineation of maxillofacial anatomy in Cone-Beam Computed Tomography (CBCT) is essential for dental planning, but robust automated segmentation remains challenging due to limited public multi-structure datasets and the high computational burden of 3D deep learning models. We present and release ToothFairy3, a large-scale CBCT benchmark that extends ToothFairy2 with 102 additional fully annotated scans and an expanded taxonomy covering 77 classes, including 32 tooth-specific pulp cavities and small neurovascular structures. ToothFairy3 comprises 582 volumes (over 40 000 annotated objects), 532 released with voxel-level labels and 50 held out for leakage-free, server-side evaluation. We also introduce U-Mamba2, an efficient U-Net-style architecture that inserts a Mamba2 state-space block at the bottleneck to capture global context with favorable computational scaling. Our proposed domain-informed training further improves the learning of maxillofacial anatomies. Across CNN, Transformer, and Mamba baselines, U-Mamba2 achieves competitive Dice/HD95 scores with lower latency and, compared with training on state-of-the-art public CBCT datasets, ToothFairy3-trained models generalize best to the hidden test set, particularly for maxillary structures.

Links

Paper record: IRIS.

Dataset: ToothFairy3 and Grand Challenge evaluation.

Code: U-Mamba2.

The project builds on ToothFairy2 by increasing anatomical granularity and field coverage. Its main contribution is not only a larger dataset, but a more clinically relevant benchmark for small canals, tooth-specific pulp cavities, and upper-jaw structures.

---

← Back to projects