Longitudinal Progression Prediction of Alzheimer's Disease with Tabular Foundation Model

2025-08-25

AI Summary

Predicting Alzheimer’s disease (AD) progression remains highly challenging due to its multifactorial etiology and the difficulty of modeling multimodal longitudinal clinical data. To address this, we propose L2C-TabPFN—a novel framework that first transforms longitudinal clinical records into static tabular features via a “Longitudinal-to-Contextual” (L2C) encoding scheme, then leverages the pre-trained tabular foundation model TabPFN to jointly predict diagnostic labels, cognitive scores (e.g., CDR-SB, MMSE), and key neuroimaging biomarkers—specifically, ventricular volume. Evaluated on the TADPOLE benchmark, L2C-TabPFN achieves state-of-the-art performance in ventricular volume prediction and competitive accuracy in diagnosis and cognitive score forecasting. This work demonstrates, for the first time, the efficacy of tabular foundation models in modeling long-term neurodegenerative disease progression, establishing a new paradigm for precision AD prognosis.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Alzheimer's disease is a progressive neurodegenerative disorder that remains challenging to predict due to its multifactorial etiology and the complexity of multimodal clinical data. Accurate forecasting of clinically relevant biomarkers, including diagnostic and quantitative measures, is essential for effective monitoring of disease progression. This work introduces L2C-TabPFN, a method that integrates a longitudinal-to-cross-sectional (L2C) transformation with a pre-trained Tabular Foundation Model (TabPFN) to predict Alzheimer's disease outcomes using the TADPOLE dataset. L2C-TabPFN converts sequential patient records into fixed-length feature vectors, enabling robust prediction of diagnosis, cognitive scores, and ventricular volume. Experimental results demonstrate that, while L2C-TabPFN achieves competitive performance on diagnostic and cognitive outcomes, it provides state-of-the-art results in ventricular volume prediction. This key imaging biomarker reflects neurodegeneration and progression in Alzheimer's disease. These findings highlight the potential of tabular foundational models for advancing longitudinal prediction of clinically relevant imaging markers in Alzheimer's disease.

Problem

• Predicting Alzheimer's disease progression using multimodal clinical data
• Forecasting diagnostic outcomes and quantitative biomarker measures
• Integrating longitudinal patient records into fixed-length feature vectors

Innovation

• L2C-TabPFN integrates longitudinal-to-cross-sectional transformation
• Uses pre-trained Tabular Foundation Model for prediction
• Converts sequential records into fixed-length feature vectors