ARMARecon: An ARMA Convolutional Filter based Graph Neural Network for Neurodegenerative Dementias Classification

TL;DR

ARMARecon is a graph neural network framework that combines ARMA filtering and reconstruction objectives to improve early classification of neurodegenerative dementias using white-matter imaging data.

Contribution

It introduces a novel ARMA-based graph filtering approach integrated with a reconstruction-driven loss for enhanced disease classification.

Findings

Outperforms state-of-the-art methods on ADNI and NIFD datasets.

Effectively models local and global white-matter connectivity.

Mitigates over-smoothing in graph neural networks.

Abstract

Early detection of neurodegenerative diseases such as Alzheimer's Disease (AD) and Frontotemporal Dementia (FTD) is essential for reducing the risk of progression to severe disease stages. As AD and FTD propagate along white-matter regions in a global, graph-dependent manner, graph-based neural networks are well suited to capture these patterns. Hence, we introduce ARMARecon, a unified graph learning framework that integrates Autoregressive Moving Average (ARMA) graph filtering with a reconstruction-driven objective to enhance feature representation and improve classification accuracy. ARMARecon effectively models both local and global connectivity by leveraging 20-bin Fractional Anisotropy (FA) histogram features extracted from white-matter regions, while mitigating over-smoothing. Overall, ARMARecon achieves superior performance compared to state-of-the-art methods on the multi-site dMRI datasets ADNI and NIFD.