Search papers, labs, and topics across Lattice.
The paper introduces MATHENA, a unified Mamba-based framework for dental diagnosis from Orthopantomograms (OPGs), addressing tooth detection, caries segmentation, anomaly detection, and dental developmental staging. MATHENA uses a multi-resolution SSM-driven detector (MATHE) for efficient global context modeling and a Mamba-UNet (HENA) with a triple-head architecture for the downstream tasks. Experiments on a new benchmark dataset, PARTHENON, demonstrate that MATHENA achieves state-of-the-art results across all four diagnostic tasks.
Mamba's linear-complexity SSMs can unify and accelerate dental diagnosis, achieving state-of-the-art results in tooth detection, caries segmentation, anomaly detection, and developmental staging.
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.
NVIDIA