NEUBORN: The Neurodevelopmental Evolution framework Using BiOmechanical RemodelliNg

2025-08-13

AI Summary

Traditional group-average cortical modeling frameworks fail to capture fine-grained developmental features, limiting individualized neurodevelopmental analysis and early detection of atypical development. To address this, we propose the first biologically informed, hierarchical deep learning framework for personalized cortical growth modeling, integrating biomechanical constraints with differentiable diffeomorphic registration and neural networks to jointly learn longitudinal deformation fields. Our approach ensures physiologically plausible, smooth, and interpretable individual growth trajectories. Evaluated on the Developing Human Connectome Project (dHCP) neonatal MRI dataset, the method significantly reduces the incidence of negative Jacobian determinants—indicating improved diffeomorphism—and yields smoother, more anatomically consistent deformation fields. It also achieves superior alignment with population-level developmental trends. Critically, it enhances sensitivity to early deviations in brain maturation, enabling more robust identification of atypical neurodevelopmental trajectories.

Technology Category

• Machine Learning: Learning with Manifolds
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: User model development and evaluation
• 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

Understanding individual cortical development is essential for identifying deviations linked to neurodevelopmental disorders. However, current normative modelling frameworks struggle to capture fine-scale anatomical details due to their reliance on modelling data within a population-average reference space. Here, we present a novel framework for learning individual growth trajectories from biomechanically constrained, longitudinal, diffeomorphic image registration, implemented via a hierarchical network architecture. Trained on neonatal MRI data from the Developing Human Connectome Project, the method improves the biological plausibility of warps, generating growth trajectories that better follow population-level trends while generating smoother warps, with fewer negative Jacobians, relative to state-of-the-art baselines. The resulting subject-specific deformations provide interpretable, biologically grounded mappings of development. This framework opens new possibilities for predictive modeling of brain maturation and early identification of malformations of cortical development.

Problem

• Modeling individual cortical growth trajectories accurately
• Capturing fine-scale anatomical details in neurodevelopment
• Improving biological plausibility of brain deformation models

Innovation

• Biomechanically constrained diffeomorphic image registration
• Hierarchical network architecture for growth trajectories
• Improved biological plausibility with smoother warps