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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Abburi, VSS Tejaswi, Singhal, Ananya, Shigwan, Saurabh J., Kumar, Nitin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908773181816832
author Abburi, VSS Tejaswi
Singhal, Ananya
Shigwan, Saurabh J.
Kumar, Nitin
author_facet Abburi, VSS Tejaswi
Singhal, Ananya
Shigwan, Saurabh J.
Kumar, Nitin
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12067
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ARMARecon: An ARMA Convolutional Filter based Graph Neural Network for Neurodegenerative Dementias Classification
Abburi, VSS Tejaswi
Singhal, Ananya
Shigwan, Saurabh J.
Kumar, Nitin
Computer Vision and Pattern Recognition
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.
title ARMARecon: An ARMA Convolutional Filter based Graph Neural Network for Neurodegenerative Dementias Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.12067