CognitiveTwin: Robust Multi-Modal Digital Twins for Predicting Cognitive Decline in Alzheimer's Disease

2026-04-24

AI Summary

This study addresses the significant inter-individual variability in cognitive decline among Alzheimer’s disease patients, a challenge that existing methods struggle to reconcile with respect to prediction accuracy, fairness across demographic groups, and robustness to missing data. To this end, the work introduces— for the first time—the digital twin paradigm into this domain, proposing a multimodal framework that integrates longitudinal cognitive scores, MRI, PET, cerebrospinal fluid biomarkers, and genetic data. Cross-modal fusion is achieved via Transformers, while temporal dynamics are modeled using a deep Markov model to enable personalized cognitive trajectory forecasting. Evaluated on 1,666 TADPOLE participants, the approach demonstrates high predictive accuracy alongside strong fairness across populations and robustness to non-randomly missing data, offering potential utility in clinical trial enrichment and individualized care planning.

Technology Category

• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Algorithmic accountability and transparency on the web
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Predicting individual cognitive decline in Alzheimer's disease (AD) is difficult due to the heterogeneity of disease progression. Reliable clinical tools require not only high accuracy but also fairness across demographics and robustness to missing data. We present CognitiveTwin, a digital twin framework that predicts patient-specific cognitive trajectories. The model integrates multi-modal longitudinal data (cognitive scores, magnetic resonance imaging, positron emission tomography, cerebrospinal fluid biomarkers, and genetics). We use a Transformer-based architecture to fuse these modalities and a Deep Markov Model to capture temporal dynamics. We trained and evaluated the framework using data from 1,666 patients in the TADPOLE (Alzheimer's Disease Neuroimaging Initiative) dataset. We assessed the model for prediction error, demographic fairness, and robustness to missing-not-at-random (MNAR) data patterns. ognitiveTwin provides accurate and personalized predictions of cognitive decline. Its demonstrated fairness across patient demographics and resilience to clinical dropout make it a reliable tool for clinical trial enrichment and personalized care planning.

Problem

• Alzheimer's disease
• cognitive decline prediction
• demographic fairness
• missing data robustness
• digital twin

Innovation

• Digital Twin
• Multi-Modal Fusion
• Transformer Architecture
• Deep Markov Model
• Missing-Not-at-Random (MNAR) Robustness