OrthoAI: A Neurosymbolic Framework for Evidence-Grounded Biomechanical Reasoning in Clear Aligner Orthodontics

TL;DR

OrthoAI introduces a neurosymbolic framework that combines sparse-supervision segmentation, knowledge-grounded biomechanical constraints, and multi-criteria treatment evaluation to improve automated decision support in clear aligner orthodontics.

Contribution

The paper presents a novel neurosymbolic approach integrating geometric perception and clinical reasoning for orthodontic treatment planning, with new methods for sparse annotation, constraint inference, and treatment scoring.

Findings

Segmentation achieves 81.4% Tooth Identification Rate.

End-to-end inference runs in under 4 seconds on CPU.

Framework demonstrates potential for clinical deployment.

Abstract

Automated clinical decision support for clear aligner orthodontics faces a key challenge: bridging geometric perception (3D tooth segmentation) with clinical reasoning (biomechanical feasibility). We address this with OrthOAI, introducing three methodological contributions. First, sparse-supervision segmentation: a landmark-to-point-cloud synthesis protocol enables training from sparse anatomical annotations (6-8 points per tooth) instead of dense labels, combined with a clinically stratified loss mixing label-smoothed cross-entropy and a batch-adaptive Dice term for class imbalance. Second, knowledge-grounded constraint inference: biomechanical feasibility is modeled as a Constraint Satisfaction Problem over a domain ontology of tooth movements, encoding evidence-based per-stage limits as soft and hard constraints. Third, multi-criteria treatment evaluation: treatment quality is scored through a formal Multi-Criteria Decision Analysis framework using a weighted Additive Value Function grounded in clinical priority theory. On landmark-reconstructed point clouds from 3DTeethLand (MICCAI 2024), segmentation reaches 81.4% Tooth Identification Rate with 60,705 parameters. Ablations quantify the impact of each design choice. End-to-end inference runs in under 4 seconds on CPU. We also outline the gap between the current prototype-trained on synthetic ellipsoidal approximations-and clinical deployment, with a roadmap for validation. Code and weights are released.