AGE-Net: Spectral--Spatial Fusion and Anatomical Graph Reasoning with Evidential Ordinal Regression for Knee Osteoarthritis Grading

TL;DR

This paper introduces AGE-Net, a novel framework combining spectral-spatial fusion, anatomical graph reasoning, and evidential regression to improve knee osteoarthritis grading accuracy and uncertainty estimation from radiographs.

Contribution

AGE-Net is the first to integrate spectral-spatial fusion, anatomical graph reasoning, and evidential ordinal regression for enhanced knee osteoarthritis grading.

Findings

Achieved a quadratic weighted kappa of 0.9017 on the dataset.

Outperformed baseline CNN models in accuracy and robustness.

Demonstrated effective uncertainty quantification and explainability.

Abstract

Automated Kellgren--Lawrence (KL) grading from knee radiographs is challenging due to subtle structural changes, long-range anatomical dependencies, and ambiguity near grade boundaries. We propose AGE-Net, a ConvNeXt-based framework that integrates Spectral--Spatial Fusion (SSF), Anatomical Graph Reasoning (AGR), and Differential Refinement (DFR). To capture predictive uncertainty and preserve label ordinality, AGE-Net employs a Normal-Inverse-Gamma (NIG) evidential regression head and a pairwise ordinal ranking constraint. On a knee KL dataset, AGE-Net achieves a quadratic weighted kappa (QWK) of 0.9017 +/- 0.0045 and a mean squared error (MSE) of 0.2349 +/- 0.0028 over three random seeds, outperforming strong CNN baselines and showing consistent gains in ablation studies. We further outline evaluations of uncertainty quality, robustness, and explainability, with additional experimental figures to be included in the full manuscript.