MRExtrap: Longitudinal Aging of Brain MRIs using Linear Modeling in Latent Space

2025-08-26

AI Summary

This study aims to model age-related structural dynamics in 3D brain MRI to elucidate progression patterns of neurodegenerative disorders such as Alzheimer’s disease. Method: We propose MRExtrap—a framework leveraging a convolutional autoencoder to learn a low-dimensional latent space where aging trajectories are approximately linear. It integrates a population-level prior and subject-specific priors, jointly modeling age information to perform linear extrapolation and estimate individualized latent progression rates (β). The framework supports Bayesian posterior updating with multi-timepoint scans for personalized dynamic prediction. Results: Evaluated on the ADNI dataset, MRExtrap achieves superior single-scan prediction accuracy over GAN-based baselines; β correlates significantly with established regional atrophy patterns; and multi-scan data enable robust refinement of individual aging trajectories. This work is the first to systematically demonstrate the linearity and tractability of aging in MRI-derived latent spaces, enabling interpretable, updateable, and personalized inference of neurobiological aging.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Multi-modal Vision

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Simulating aging in 3D brain MRI scans can reveal disease progression patterns in neurological disorders such as Alzheimer's disease. Current deep learning-based generative models typically approach this problem by predicting future scans from a single observed scan. We investigate modeling brain aging via linear models in the latent space of convolutional autoencoders (MRExtrap). Our approach, MRExtrap, is based on our observation that autoencoders trained on brain MRIs create latent spaces where aging trajectories appear approximately linear. We train autoencoders on brain MRIs to create latent spaces, and investigate how these latent spaces allow predicting future MRIs through linear extrapolation based on age, using an estimated latent progression rate $oldsymbolβ$. For single-scan prediction, we propose using population-averaged and subject-specific priors on linear progression rates. We also demonstrate that predictions in the presence of additional scans can be flexibly updated using Bayesian posterior sampling, providing a mechanism for subject-specific refinement. On the ADNI dataset, MRExtrap predicts aging patterns accurately and beats a GAN-based baseline for single-volume prediction of brain aging. We also demonstrate and analyze multi-scan conditioning to incorporate subject-specific progression rates. Finally, we show that the latent progression rates in MRExtrap's linear framework correlate with disease and age-based aging patterns from previously studied structural atrophy rates. MRExtrap offers a simple and robust method for the age-based generation of 3D brain MRIs, particularly valuable in scenarios with multiple longitudinal observations.

Problem

• Modeling brain MRI aging trajectories using linear latent space extrapolation
• Predicting future brain scans from single or multiple longitudinal observations
• Correlating latent progression rates with disease and structural atrophy patterns

Innovation

• Linear modeling in latent space
• Convolutional autoencoders for aging simulation
• Bayesian posterior sampling for refinement