Artificial Intelligence for Personalized Prediction of Alzheimer's Disease Progression: A Survey of Methods, Data Challenges, and Future Directions
Gulsah Hancerliogullari Koksalmis, Bulent Soykan, Laura J. Brattain, Hsin-Hsiung Huang
TL;DR
This survey addresses the challenge of predicting Alzheimer’s disease progression at the individual level by surveying AI-driven approaches that integrate multimodal, longitudinal data. It analyzes methods ranging from state-space and deep temporal models to graph neural networks and AI-driven digital twins, including generative techniques like VAEs and GANs to address data limitations. The authors synthesize insights on data resources, validation practices, and the balance between predictive performance and interpretability, and they articulate open challenges such as external validation, clinical integration, and ethical considerations. The work highlights how multimodal integration and dynamic simulation can enable personalized prognostication and more efficient clinical trials, guiding future research toward robust, generalizable, and clinically actionable AI tools for personalized AD care.
Abstract
Alzheimer's Disease (AD) is marked by significant inter-individual variability in its progression, complicating accurate prognosis and personalized care planning. This heterogeneity underscores the critical need for predictive models capable of forecasting patient-specific disease trajectories. Artificial Intelligence (AI) offers powerful tools to address this challenge by analyzing complex, multi-modal, and longitudinal patient data. This paper provides a comprehensive survey of AI methodologies applied to personalized AD progression prediction. We review key approaches including state-space models for capturing temporal dynamics, deep learning techniques like Recurrent Neural Networks for sequence modeling, Graph Neural Networks (GNNs) for leveraging network structures, and the emerging concept of AI-driven digital twins for individualized simulation. Recognizing that data limitations often impede progress, we examine common challenges such as high dimensionality, missing data, and dataset imbalance. We further discuss AI-driven mitigation strategies, with a specific focus on synthetic data generation using Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) to augment and balance datasets. The survey synthesizes the strengths and limitations of current approaches, emphasizing the trend towards multimodal integration and the persistent need for model interpretability and generalizability. Finally, we identify critical open challenges, including robust external validation, clinical integration, and ethical considerations, and outline promising future research directions such as hybrid models, causal inference, and federated learning. This review aims to consolidate current knowledge and guide future efforts in developing clinically relevant AI tools for personalized AD prognostication.