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

TL;DR

This paper presents L2C-TabPFN, a novel method combining longitudinal data transformation with a pre-trained tabular model to predict Alzheimer's disease progression and biomarkers, achieving state-of-the-art results in ventricular volume prediction.

Contribution

Introduces L2C-TabPFN, a new approach that leverages a longitudinal-to-cross-sectional transformation with a tabular foundation model for improved disease progression prediction.

Findings

State-of-the-art ventricular volume prediction accuracy.

Competitive performance in diagnosis and cognitive score prediction.

Effective transformation of longitudinal data into fixed-length features.

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.