Alzheimer's Disease Brain Network Mining

TL;DR

This paper introduces MATCH-AD, a semi-supervised learning framework that effectively diagnoses Alzheimer's disease using limited labeled data and large unlabeled datasets, significantly improving accuracy and reliability.

Contribution

The paper presents a novel semi-supervised framework combining deep learning, graph propagation, and optimal transport to enhance AD diagnosis with scarce labels.

Findings

Achieves near-perfect diagnostic accuracy with less than one-third labeled data.

Outperforms all baseline methods with almost perfect agreement.

Maintains clinical usefulness under severe label scarcity.

Abstract

Machine learning approaches for Alzheimer's disease (AD) diagnosis face a fundamental challenges. Clinical assessments are expensive and invasive, leaving ground truth labels available for only a fraction of neuroimaging datasets. We introduce Multi view Adaptive Transport Clustering for Heterogeneous Alzheimer's Disease (MATCH-AD), a semi supervised framework that integrates deep representation learning, graph-based label propagation, and optimal transport theory to address this limitation. The framework leverages manifold structure in neuroimaging data to propagate diagnostic information from limited labeled samples to larger unlabeled populations, while using Wasserstein distances to quantify disease progression between cognitive states. Evaluated on nearly five thousand subjects from the National Alzheimer's Coordinating Center, encompassing structural MRI measurements from hundreds of brain regions, cerebrospinal fluid biomarkers, and clinical variables MATCHAD achieves near-perfect diagnostic accuracy despite ground truth labels for less than one-third of subjects. The framework substantially outperforms all baseline methods, achieving kappa indicating almost perfect agreement compared to weak agreement for the best baseline, a qualitative transformation in diagnostic reliability. Performance remains clinically useful even under severe label scarcity, and we provide theoretical convergence guarantees with proven bounds on label propagation error and transport stability. These results demonstrate that principled semi-supervised learning can unlock the diagnostic potential of the vast repositories of partially annotated neuroimaging data accumulating worldwide, substantially reducing annotation burden while maintaining accuracy suitable for clinical deployment.