M$^3$AD: Multi-task Multi-gate Mixture of Experts for Alzheimer's Disease Diagnosis with Conversion Pattern Modeling

TL;DR

This paper introduces M$^3$AD, a multi-task deep learning framework that models Alzheimer's disease progression and diagnosis using structural MRI, achieving high accuracy and better generalization with fewer modalities.

Contribution

The study presents a novel multi-task mixture of experts model that jointly performs diagnosis and transition modeling, incorporating demographic priors and a specialized architecture for improved Alzheimer's disease prediction.

Findings

Achieved 95.13% accuracy in three-class classification.

Predicted cognitive transition with 97.76% accuracy.

Outperformed state-of-the-art methods with fewer modalities.

Abstract

Alzheimer's disease (AD) progression follows a complex continuum from normal cognition (NC) through mild cognitive impairment (MCI) to dementia, yet most deep learning approaches oversimplify this into discrete classification tasks. This study introduces M$^3$AD, a novel multi-task multi-gate mixture of experts framework that jointly addresses diagnostic classification and cognitive transition modeling using structural MRI. We incorporate three key innovations: (1) an open-source T1-weighted sMRI preprocessing pipeline, (2) a unified learning framework capturing NC-MCI-AD transition patterns with demographic priors (age, gender, brain volume) for improved generalization, and (3) a customized multi-gate mixture of experts architecture enabling effective multi-task learning with structural MRI alone. The framework employs specialized expert networks for diagnosis-specific pathological patterns while shared experts model common structural features across the cognitive continuum. A two-stage training protocol combines SimMIM pretraining with multi-task fine-tuning for joint optimization. Comprehensive evaluation across six datasets comprising 12,037 T1-weighted sMRI scans demonstrates superior performance: 95.13% accuracy for three-class NC-MCI-AD classification and 99.15% for binary NC-AD classification, representing improvements of 4.69% and 0.55% over state-of-the-art approaches. The multi-task formulation simultaneously achieves 97.76% accuracy in predicting cognitive transition. Our framework outperforms existing methods using fewer modalities and offers a clinically practical solution for early intervention. Code: https://github.com/csyfjiang/M3AD.