ARMA-C3: A Contrastive ARMA Convolutional Framework for Unsupervised and Semi-supervised Classification

2026-05-25

AI Summary

This work addresses the challenge of early diagnosis in biomedicine and neurodegenerative diseases, where scarce annotations and complex imaging modalities hinder performance. The authors propose a unified unsupervised and semi-supervised node classification framework based on graph learning. Their approach uniquely integrates contrastive learning with ARMA graph convolution and incorporates graph-cut regularization to model subjects as graph nodes and their interrelationships, effectively capturing subject-level dependency structures. This design enhances generalization under label scarcity and class imbalance. Evaluated on ADNI, NIFD, and three medical imaging benchmark datasets, the model significantly outperforms conventional clustering methods, mainstream machine learning approaches, and existing graph deep learning techniques, demonstrating superior representation learning and cross-modal generalization capabilities.

Abstract

In biomedical and neurodegenerative disorders, accurate and early disease identification remains challenging due to the scarcity of labeled data and the complexity of imaging patterns. To address these challenges, we introduce ARMA-C3, a unified unsupervised and semi-supervised graph learning framework for node classification based on contrastive learning and graph-cut regularization to learn structurally meaningful and discriminative representations. By modeling samples or images as graph nodes and exploiting inter-sample relationships, the proposed framework captures subject-level dependencies that conventional machine learning methods typically overlook. We conduct extensive binary classification experiments across five clinically relevant datasets: the Alzheimer's Disease Neuroimaging Initiative (ADNI), the Neuroimaging in Frontotemporal Dementia (NIFD) dataset, and three medical imaging benchmarks (BreastMNIST, PneumoniaMNIST, and a liver ultrasound dataset). Experimental results demonstrate that ARMA-C3 achieves competitive and frequently superior performance compared to classical clustering techniques, state-of-the-art machine learning models, and existing graph-based deep learning approaches across multiple evaluation settings, particularly under limited supervision and severe class imbalance. The proposed framework further demonstrates robust representation learning and strong cross-modal generalization across diverse biomedical imaging modalities.

Problem

• disease identification
• labeled data scarcity
• imaging pattern complexity
• biomedical imaging
• neurodegenerative disorders

Innovation

• contrastive learning
• ARMA graph convolution
• graph-cut regularization
• unsupervised representation learning
• cross-modal generalization