MATHENA: Mamba-based Architectural Tooth Hierarchical Estimator and Holistic Evaluation Network for Anatomy

2026-04-01

AI Summary

This work addresses the challenge of jointly performing four critical diagnostic tasks—dental detection, caries segmentation, anomaly detection, and dental age staging—from panoramic radiographs by proposing the first unified multitask framework based on the Mamba state-space model. The approach employs a four-directional visual state block to generate multiresolution tooth regions and introduces a lightweight tri-head Mamba-UNet architecture for efficient joint learning. A novel collaborative design integrates the MATHE detector and HENA assessment network, complemented by an upstream representation freezing strategy to enhance training efficiency. Evaluated on the newly established PARTHENON benchmark, the method achieves 93.78% mAP@50 in tooth detection, 90.11% Dice score for caries segmentation, 88.35% accuracy in anomaly detection, and 72.40% accuracy in dental age staging.

Abstract

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.

Problem

• tooth detection
• caries segmentation
• anomaly detection
• dental developmental staging
• orthopantomogram

Innovation

• Mamba
• State Space Model
• Multi-task Dental Diagnosis
• Vision State Space
• Linear-complexity Context Modeling