Forecasting Future Anatomies: Longitudianl Brain Mri-to-Mri Prediction

2025-11-04

AI Summary

This study addresses the challenge of fine-grained, individualized modeling of neurodegenerative spatial dynamics by predicting whole-brain anatomical structure years into the future from baseline MRI. Method: We propose the first end-to-end whole-brain MRI-to-MRI longitudinal prediction framework, directly modeling voxel-wise morphological change across years to inherently capture complex, nonlinear degeneration patterns. Five deep architectures—UNet, U2-Net, UNETR, Time-Embedding UNet, and ODE-UNet—are systematically evaluated on ADNI and AIBL cohorts. Contribution/Results: The best-performing model significantly outperforms baselines in voxel-level similarity (SSIM/PSNR) and clinical-region change consistency. Cross-cohort generalization experiments demonstrate robustness, with high predictive fidelity maintained on independent external datasets. The framework delivers an interpretable, deployable computational imaging biomarker for personalized brain health prognosis and early intervention.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Diffusion Models for Vision

Application Category

• 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
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Predicting future brain state from a baseline magnetic resonance image (MRI) is a central challenge in neuroimaging and has important implications for studying neurodegenerative diseases such as Alzheimer's disease (AD). Most existing approaches predict future cognitive scores or clinical outcomes, such as conversion from mild cognitive impairment to dementia. Instead, here we investigate longitudinal MRI image-to-image prediction that forecasts a participant's entire brain MRI several years into the future, intrinsically modeling complex, spatially distributed neurodegenerative patterns. We implement and evaluate five deep learning architectures (UNet, U2-Net, UNETR, Time-Embedding UNet, and ODE-UNet) on two longitudinal cohorts (ADNI and AIBL). Predicted follow-up MRIs are directly compared with the actual follow-up scans using metrics that capture global similarity and local differences. The best performing models achieve high-fidelity predictions, and all models generalize well to an independent external dataset, demonstrating robust cross-cohort performance. Our results indicate that deep learning can reliably predict participant-specific brain MRI at the voxel level, offering new opportunities for individualized prognosis.

Problem

• Predicting future brain MRI scans from baseline neuroimaging data
• Modeling neurodegenerative disease progression through longitudinal image prediction
• Evaluating deep learning architectures for voxel-level brain anatomy forecasting

Innovation

• Deep learning architectures predict future brain MRIs
• Models generalize across multiple longitudinal patient cohorts
• Voxel-level MRI forecasting enables individualized neurodegeneration prognosis