Tracing Like a Clinician: Anatomy-Guided Spatial Priors for Cephalometric Landmark Detection

Abstract

Clinicians trace cephalometric radiographs following a structured anatomical workflow, yet no prior system encodes this into computation. We present a five-phase anatomy-guided pipeline producing confidence-weighted spatial priors that shape HRNet-W32 training, achieving 1.04 mm mean radial error on 25 landmarks across 1,502 radiographs from 7+ imaging devices. A training x inference prior matrix isolates the mechanism: anatomical priors maintain a 1% validation-to-test gap versus 88% without priors (1.94 mm), despite identical validation convergence. The matrix establishes that all trained models are inference-independent, the expanded architecture alone provides no benefit, random priors yield partial but unstable improvement (1.72 mm), and only image-specific anatomically correct priors produce the 1.04 mm result -- functioning as a training-time regularizer requiring no automated prior generation at deployment. Five-fold cross-validation (p=0.0015), patient-level permutation testing (p<0.0001, n=151), quantified Grad-CAM analysis (88% vs. 74% in-zone activation, p<0.001), and clinical measurement validation (skeletal classification kappa=0.79-0.84, zero Class II<->III reversals, ICC>0.95) provide converging evidence. Cross-domain experiments on echocardiography, cervical spine, and hand radiography support the hypothesis that prior effectiveness scales with the spatial entropy of the landmark distribution.

v3: 21 pages, 15 tables, 12 figures + supplementary materials (8 tables, 3 figures). v4: quantified Grad-CAM analysis (Table 13), corrected clinical measurements (Table 6: bias, MAE, ICC; vertical kappa 1.00->0.78), reviewer wording fixes. Code and weights: https://github.com/sidwiz/cephtrace-research, https://huggingface.co/CephTrace/cephtrace-v4