Dynamic causal discovery in Alzheimer's disease through latent pseudotime modelling

TL;DR

This paper introduces a method for causal discovery in Alzheimer's disease by modeling a latent pseudotime trajectory, capturing disease progression dynamics that improve diagnosis prediction and reveal evolving biomarker interactions.

Contribution

It applies a latent variable model to infer disease pseudotime from real-world data, enabling dynamic causal analysis beyond static assumptions.

Findings

Pseudotime predicts diagnosis with AUC 0.82, outperforming age.

Incorporating background knowledge improves graph accuracy.

Reveals evolving interactions among AD biomarkers.

Abstract

The application of causal discovery to diseases like Alzheimer's (AD) is limited by the static graph assumptions of most methods; such models cannot account for an evolving pathophysiology, modulated by a latent disease pseudotime. We propose to apply an existing latent variable model to real-world AD data, inferring a pseudotime that orders patients along a data-driven disease trajectory independent of chronological age, then learning how causal relationships evolve. Pseudotime outperformed age in predicting diagnosis (AUC 0.82 vs 0.59). Incorporating minimal, disease-agnostic background knowledge substantially improved graph accuracy and orientation. Our framework reveals dynamic interactions between novel (NfL, GFAP) and established AD markers, enabling practical causal discovery despite violated assumptions.