Predicting Alzheimer's disease progression using rs-fMRI and a history-aware graph neural network

Mahdi Moghaddami; Mohammad-Reza Siadat; Austin Toma; Connor Laming; Huirong Fu

cs.CVApr 7, 2026

Predicting Alzheimer's disease progression using rs-fMRI and a history-aware graph neural network

Authors: Mahdi Moghaddami, Mohammad-Reza Siadat, Austin Toma, Connor Laming, Huirong Fu

Organizations: Oakland University, 502 Meadow Brook Road, Rochester, MI 48309, USA

Abstract

Alzheimer's disease (AD) is a neurodegenerative disorder that affects more than seven million people in the United States alone. AD currently has no cure, but there are ways to potentially slow its progression if caught early enough. In this study, we propose a graph neural network (GNN)-based model for predicting whether a subject will transition to a more severe stage of cognitive impairment at their next clinical visit. We consider three stages of cognitive impairment in order of severity: cognitively normal (CN), mild cognitive impairment (MCI), and AD. We use functional connectivity graphs derived from resting-state functional magnetic resonance imaging (rs-fMRI) scans of 303 subjects, each with a different number of visits. Our GNN-based model incorporates a recurrent neural network (RNN) block, enabling it to process data from the subject's entire visit history. It can also work with irregular time gaps between visits by incorporating visit distance information into our input features. Our model demonstrates robust predictive performance, even with missing visits in the subjects' visit histories. It achieves an accuracy of 82.9%, with an especially impressive accuracy of 68.8% on CN to MCI conversions - a task that poses a substantial challenge in the field. Our results highlight the effectiveness of rs-fMRI in predicting the onset of MCI or AD and, in conjunction with other modalities, could offer a viable method for enabling timely interventions to slow the progression of cognitive impairment.

Figures & tables

Panel of Figure 1 — Figure 1: Distribution of subjects based on their number of available rs-fMRI scans. The blue bars represent stable subjects, while the orange bars represent converter subjects.

Panel of Figure 2 — Figure 2: Distribution of subjects based on their diagnosis in their last visit

Panel of Figure 3 — Figure 3: Overview of our workflow, from raw rs-fMRI scans to the binary prediction

Panel of Figure 4 — Figure 4: Architecture of our HA-GNN model

Panel of Figure 5 — Figure 5: Architecture of the GCN pretraining model

Model	Acc.	AUC-ROC	BA	CN to MCI	MCI to AD
HA-GNN (LSTM)	0.829 $\pm$ 0.058	0.852 $\pm$ 0.065	0.771 $\pm$ 0.114	0.688	0.676
HA-GNN (RNN)	0.743 $\pm$ 0.038	0.771 $\pm$ 0.067	0.651 $\pm$ 0.114	0.5	0.514
HA-GNN (GRU)	0.733 $\pm$ 0.049	0.786 $\pm$ 0.062	0.704 $\pm$ 0.071	0.625	0.676
w/o pretraining	0.723 $\pm$ 0.079	0.642 $\pm$ 0.095	0.594 $\pm$ 0.061	0.375	0.405

Table 1: Classification results of the proposed model

Alzheimer's Disease Progression Prediction

Alzheimer’s disease progression prediction estimates future cognitive decline, diagnostic transitions, and brain changes from patient data over time.

See all 38 papers

Explainable Graph-theoretical Machine Learning with Application to Alzheimer's Disease Prediction
Mar 20, 2025Narmina Baghirova, Duy-Thanh Vũ, Duy-Cat Can +6Clinical Outcome Prediction Neuroimaging
Department of Informatics, University of Fribourg, Fribourg, Switzerland · Platform of Bioinformatics, Lausanne University Hospital (CHUV), Lausanne, Switzerland · Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland
SDE-Driven Spatio-Temporal Hypergraph Neural Networks for Irregular Longitudinal fMRI Connectome Modeling in Alzheimer's Disease
Mar 20, 2026Ruiying Chen, Yutong Wang, Houliang Zhou +3Irregular Time-Series Modeling Alzheimer's Disease
Lehigh University, Bethlehem, PA, USA · New York University, New York, NY, USA · University of Pennsylvania, Philadelphia, PA, USA
A Two-Scan Deep Learning Model for Predicting Dementia in Mild Cognitive Impairment
May 27, 2026Sara Fin, Alireza Moayedikia, David J. White +2Hybrid CNN-Transformer Architectures Longitudinal Medical Image Analysis
Australian Regenerative Medicine Institute, Monash University, Melbourne, VIC, Australia · School of Business Law and Entrepreneurship, Swinburne University of Technology, Melbourne, VIC, Australia · department of Business Technology and Entrepreneurship of Swinburne University of Technology, John Street, Hawthorn 3122, VIC, Australia +2

Mar 20, 2025J/K move · Enter open · S save

Explainable Graph-theoretical Machine Learning with Application to Alzheimer's Disease Prediction

Narmina Baghirova, Duy-Thanh Vũ, Duy-Cat Can, Christelle Schneuwly Diaz+5

Department of Informatics, University of Fribourg, Fribourg, Switzerland · Platform of Bioinformatics, Lausanne University Hospital (CHUV), Lausanne, Switzerland · Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland

Clinical Outcome Prediction Neuroimaging

Open paper AskPDF arXiv

Dementia affects over 55 million people worldwide, projected to reach 139 million by 2050, with Alzheimer's disease (AD) accounting for 60-70% of cases. AD is associated with disruptions in metabolic brain connectivity. Detecting these disruptions early is crucial for AD management. FDG-PET is a useful tool for identifying such impairments. However, most studies rely on group-level analyses or thresholding, potentially masking individual differences and overlooking weaker yet biologically critical brain connections. Moreover, AD prediction largely focuses on univariate rather than multivariate outcomes. To address this, we introduce explainable graph-theoretical machine learning (XGML), a framework for constructing individual metabolic brain graphs and identifying subgraphs most predictive of multivariate disease-related outcomes. Using Alzheimer's Disease Neuroimaging Initiative (ADNI) FDG-PET data, we compared six graph representations against three non-graph baselines, each with six machine learning models using repeated stratified 3-fold cross-validation (10 repeats). The best configuration combined kernel density estimation with Hellinger distance and random forest. Across eight cognitive scores, it reached an overall Fisher-z-averaged Pearson correlation of r=0.595, with strongest performance for ADAS13 (r=0.67), ADAS11 (r=0.65), and ADASQ4 (r=0.62). We identified key edges that were jointly but differentially predictive across outcomes, suggesting their potential as network biomarkers of cognitive decline. Preliminary external feasibility validation on an OASIS3 cohort yielded weak predictive performance for CDRSB (r=0.26) and MMSE (r=0.18), likely reflecting cohort, protocol, and diagnostic differences. Overall, our results suggest the promise of graph-theoretical machine learning for biomarker discovery, disease prediction, and understanding the neural mechanisms underlying AD.

Abstract

Alzheimer's Disease Progression Prediction

Explore similar work

Explainable Graph-theoretical Machine Learning with Application to Alzheimer's Disease Prediction

SDE-Driven Spatio-Temporal Hypergraph Neural Networks for Irregular Longitudinal fMRI Connectome Modeling in Alzheimer's Disease

A Two-Scan Deep Learning Model for Predicting Dementia in Mild Cognitive Impairment