Dental diagnosis from Orthopantomograms (OPGs) requires coordination of tooth detection, caries segmentation (CarSeg), anomaly detection (AD), and dental developmental staging (DDS). We propose Mamba-based Architectural Tooth Hierarchical Estimator and Holistic Evaluation Network for Anatomy (MATHENA), a unified framework leveraging Mamba's linear-complexity State Space Models (SSM) to address all four tasks. MATHENA integrates MATHE, a multi-resolution SSM-driven detector with four-directional Vision State Space (VSS) blocks for O(N) global context modeling, generating per-tooth crops. These crops are processed by HENA, a lightweight Mamba-UNet with a triple-head architecture and Global Context State Token (GCST). In the triple-head architecture, CarSeg is first trained as an upstream task to establish shared representations, which are then frozen and reused for downstream AD fine-tuning and DDS classification via linear probing, enabling stable, efficient learning. We also curate PARTHENON, a benchmark comprising 15,062 annotated instances from ten datasets. MATHENA achieves 93.78% mAP@50 in tooth detection, 90.11% Dice for CarSeg, 88.35% for AD, and 72.40% ACC for DDS.
@article{arxiv.2604.00537,
title = {MATHENA: Mamba-based Architectural Tooth Hierarchical Estimator and Holistic Evaluation Network for Anatomy},
author = {Kyeonghun Kim and Jaehyung Park and Youngung Han and Anna Jung and Seongbin Park and Sumin Lee and Jiwon Yang and Jiyoon Han and Subeen Lee and Junsu Lim and Hyunsu Go and Eunseob Choi and Hyeonseok Jung and Soo Yong Kim and Woo Kyoung Jeong and Won Jae Lee and Pa Hong and Hyuk-Jae Lee and Ken Ying-Kai Liao and Nam-Joon Kim},
journal= {arXiv preprint arXiv:2604.00537},
year = {2026}
}