Individualized multi-horizon MRI trajectory prediction for Alzheimer's Disease

TL;DR

This paper introduces a novel conditional variational autoencoder model that predicts individualized MRI trajectories for Alzheimer's disease, leveraging patient-specific data to generate high-resolution, personalized future brain images up to 10 years ahead.

Contribution

The study presents a new architecture for personalized MRI prediction using variational autoencoders, enabling extrapolation beyond existing datasets and improving specificity in Alzheimer's diagnosis.

Findings

Model produces more individualized, higher-resolution MRI predictions.

Able to extrapolate MRI trajectories up to 10 years.

Demonstrates potential for early diagnosis and treatment monitoring.

Abstract

Neurodegeneration as measured through magnetic resonance imaging (MRI) is recognized as a potential biomarker for diagnosing Alzheimer's disease (AD), but is generally considered less specific than amyloid or tau based biomarkers. Due to a large amount of variability in brain anatomy between different individuals, we hypothesize that leveraging MRI time series can help improve specificity, by treating each patient as their own baseline. Here we turn to conditional variational autoencoders to generate individualized MRI predictions given the subject's age, disease status and one previous scan. Using serial imaging data from the Alzheimer's Disease Neuroimaging Initiative, we train a novel architecture to build a latent space distribution which can be sampled from to generate future predictions of changing anatomy. This enables us to extrapolate beyond the dataset and predict MRIs up to 10 years. We evaluated the model on a held-out set from ADNI and an independent dataset (from Open Access Series of Imaging Studies). By comparing to several alternatives, we show that our model produces more individualized images with higher resolution. Further, if an individual already has a follow-up MRI, we demonstrate a usage of our model to compute a likelihood ratio classifier for disease status. In practice, the model may be able to assist in early diagnosis of AD and provide a counterfactual baseline trajectory for treatment effect estimation. Furthermore, it generates a synthetic dataset that can potentially be used for downstream tasks such as anomaly detection and classification.