ENC-ODE: Event-level Neurodegenerative Modeling in Continuous Time with Neural ODEs

2026-06-29

AI Summary

This work addresses the challenge of sparse and irregularly sampled longitudinal biomarker data in neurodegenerative diseases such as Alzheimer’s disease by proposing an event-level continuous-time modeling framework based on Neural Ordinary Differential Equations (Neural ODEs). The approach leverages diagnosis-conditioned dynamics and a goal-oriented attention mechanism to directly aggregate multimodal event information for predicting biomarker states at arbitrary time points, without compressing historical trajectories into fixed representations. Experiments on the ADNI dataset demonstrate that the proposed model significantly outperforms prevailing sequence-based models in prediction accuracy while preserving neuroscientific interpretability. By integrating mechanistic modeling with flexible attention, this method offers clinicians a scalable, interpretable, and biologically grounded predictive tool for tracking disease progression.

Abstract

Accurately predicting the temporal evolution of clinical biomarkers is crucial for the early diagnosis and management of neurodegenerative diseases such as Alzheimer's disease. However, this relies on longitudinal data to capture biomarker changes over time, which is often sparse and irregular due to the high cost, labor-intensive nature, and patient burden. To address these challenges, we propose ENC-ODE, an Event-level Neurodegenerative modeling in Continuous time with neural Ordinary Differential Equations. ENC-ODE predicts future biomarker evolution by modeling clinical events through diagnosis-conditioned continuous dynamics. A target-conditioned attention mechanism weights and aggregates event-level predictions for the target time and modality without history compression. Extensive experiments on Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset demonstrate that ENC-ODE outperforms representative sequence models while offering a scalable and neuroscientifically grounded solution for clinical support. The code is available at https://github.com/JardinDelSol/enc-ode.

Problem

• neurodegenerative diseases
• biomarker prediction
• longitudinal data
• Alzheimer's disease
• temporal evolution

Innovation

• Neural ODEs
• Event-level modeling
• Continuous-time dynamics
• Target-conditioned attention
• Neurodegenerative disease prediction