Synthesizing Individualized Aging Brains in Health and Disease with Generative Models and Parallel Transport

2025-02-28

AI Summary

This study addresses the challenge of integrating individual anatomical specificity with population-level neurobiological aging and Alzheimer’s disease (AD) progression patterns. To this end, we propose a differentiable, subject-specific brain aging and AD progression simulation framework grounded in a single baseline T1-weighted MRI scan. Our method uniquely combines a deep generative template network with a parallel transport algorithm, enabling precise adaptation of population-derived neuroimaging trajectories to subject-specific longitudinal evolution. It integrates diffeomorphic registration, geometric transfer, and a multi-scale U-Net to synthesize high-resolution, topologically consistent 3D longitudinal MRI sequences. Quantitative and qualitative evaluations on the OASIS-3 dataset confirm its efficacy in modeling both healthy aging and early AD transition. External validation further demonstrates strong generalizability across independent cohorts. The implementation is publicly available as open-source software.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Humans and AI: Brain-Sensing and Analysis

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Simulating prospective magnetic resonance imaging (MRI) scans from a given individual brain image is challenging, as it requires accounting for canonical changes in aging and/or disease progression while also considering the individual brain's current status and unique characteristics. While current deep generative models can produce high-resolution anatomically accurate templates for population-wide studies, their ability to predict future aging trajectories for individuals remains limited, particularly in capturing subject-specific neuroanatomical variations over time. In this study, we introduce Individualized Brain Synthesis (InBrainSyn), a framework for synthesizing high-resolution subject-specific longitudinal MRI scans that simulate neurodegeneration in both Alzheimer's disease (AD) and normal aging. InBrainSyn uses a parallel transport algorithm to adapt the population-level aging trajectories learned by a generative deep template network, enabling individualized aging synthesis. As InBrainSyn uses diffeomorphic transformations to simulate aging, the synthesized images are topologically consistent with the original anatomy by design. We evaluated InBrainSyn both quantitatively and qualitatively on AD and healthy control cohorts from the Open Access Series of Imaging Studies - version 3 dataset. Experimentally, InBrainSyn can also model neuroanatomical transitions between normal aging and AD. An evaluation of an external set supports its generalizability. Overall, with only a single baseline scan, InBrainSyn synthesizes realistic 3D spatiotemporal T1w MRI scans, producing personalized longitudinal aging trajectories. The code for InBrainSyn is available at: https://github.com/Fjr9516/InBrainSyn.

Problem

• Simulate individualized aging brain MRI scans
• Predict subject-specific neuroanatomical changes over time
• Model transitions between normal aging and Alzheimer's disease

Innovation

• Generative models for individualized brain aging synthesis
• Parallel transport algorithm for personalized MRI predictions
• Diffeomorphic transformations ensure anatomical consistency