Transition-Based Digital Twin Modelling for Alzheimer's Disease under Sparse Longitudinal Data

2026-06-08

AI Summary

Alzheimer’s disease progression exhibits high heterogeneity and is characterized by sparse, irregular clinical observations, which hinder personalized dynamic modeling and uncertainty quantification. To address this challenge, this work proposes a state-transition-based personalized digital twin framework that integrates multimodal longitudinal data to model clinical state transitions between consecutive visits. This approach enables accurate prediction of cognitive scores, diagnostic classification, and counterfactual trajectory simulation. By explicitly capturing state dynamics, the method demonstrates superior efficiency and robustness over conventional sequence models under sparse and irregular observation regimes, while supporting patient-specific scenario analysis with calibrated uncertainty. Experiments on the ADNI dataset confirm the framework’s strong performance in both prediction and classification tasks, underscoring the advantages and practicality of the state-transition modeling paradigm for longitudinal neurodegenerative disease studies.

Abstract

Alzheimer's disease (AD) progression is highly heterogeneous and is typically observed through sparse and irregular longitudinal data, posing challenges for prediction and personalised monitoring. Existing machine learning approaches have improved AD prediction using multimodal data, yet often focus on static classification or cohort-level risk estimation, providing limited support for subject-specific modelling and uncertainty-aware reasoning. To address these limitations, we present a personalised digital twin framework for AD prediction and scenario-based analysis using multimodal longitudinal data. The proposed approach integrates complementary modelling strategies to capture clinical transitions and temporal dependencies across visits. Using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), including cognitive assessments, clinical variables, and MRI-derived phenotypes, the framework predicts cognitive status and diagnostic categories while quantifying predictive uncertainty and enabling patient-specific what-if trajectory analysis. Evaluation on leak-free subject-level splits demonstrates strong performance in score forecasting and diagnosis classification. In this sparse and irregular ADNI setting, transition-based modelling of adjacent visits achieved higher predictive accuracy than the sequence-based branch, suggesting that local transition modelling may be more data-efficient. While sequence models remain valuable for uncertainty-aware trajectory forecasting, local transition modelling offers a more data-efficient and robust predictive strategy. These findings highlight the importance of aligning temporal modelling strategies with clinical data structure and suggest that transition-based digital twin formulations may provide a practical and interpretable approach for personalised disease forecasting in neurodegenerative disorders.

Problem

• Alzheimer's disease
• sparse longitudinal data
• personalised modelling
• heterogeneous progression
• digital twin

Innovation

• transition-based modeling
• digital twin
• Alzheimer's disease
• sparse longitudinal data
• personalized forecasting