LNODE: latent dynamics reveal the shared spatiotemporal structure of amyloid-$\beta$ progression

TL;DR

LNODE is a neural ODE-based model that captures the spatiotemporal progression of amyloid-beta in Alzheimer's disease, enabling accurate prediction and revealing disease subtypes.

Contribution

It introduces a parsimonious, cohort-shared latent neural ODE model for amyloid-beta dynamics, demonstrating high predictive accuracy and interpretability across large datasets.

Findings

Achieves over 99% R^2 in predicting amyloid PET signals.

Accurately forecasts long-term follow-up scans.

Identifies distinct Alzheimer's disease progression subtypes.

Abstract

We introduce LNODE, a mechanism-based phenomenological model for amyloid beta (A$\beta$) dynamics, calibrated using positron emission tomography (PET) imaging. A$\beta$ is a key biomarker of Alzheimer's disease. LNODE is designed to support the fusion, harmonization, quantitative analysis, and interpretation of Abeta PET scans. We evaluate LNODE on 1461 subjects in the ADNI cohort and 1070 subjects in the A4 Study, using MUSE and DKT anatomical atlases. LNODE is formulated as a regional neural ordinary differential equation (ODE) model that is jointly calibrated on all available scans within a cohort. The model captures the spatial propagation, proliferation, and clearance of A$\beta$ and incorporates a latent-state representation that modulates A$\beta$ dynamics. The temporal evolution of these latent states is governed by cohort-shared parameters, enabling LNODE to represent both population-level trajectories and subject-specific deviations. The proposed model demonstrates strong parameter identifiability and stability properties, supported by synthetic experiments and analytical analysis of the Hessian condition number. To mitigate overfitting and reduce spurious correlations, LNODE is intentionally underparameterized, employing approximately five to ten parameters per subject. Despite this parsimonious parameterization, LNODE achieves $R^2 > 0.99$ in both the ADNI and A4 datasets. LNODE exhibits strong predictive performance: in the A4 cohort, it accurately forecasts the A$\beta$ PET signal in previously unseen follow-up scans, including cases with inter-scan intervals exceeding four years. Clustering in the learned latent-state space reveals distinct subgroups, consistent with the existence of different subtypes of Alzheimer's disease progression.