Multiscale Causal Geometric Deep Learning for Modeling Brain Structure

TL;DR

This paper introduces a novel multiscale geometric deep learning framework using spectral graph theory and disentangled learning to improve brain structure modeling from multimodal MRI data, enhancing interpretability and robustness.

Contribution

It presents a new approach combining Laplacian harmonics, spectral graph attentions, and disentangled variational autoencoders for multi-scale brain analysis, with causal factor separation.

Findings

Outperforms existing models in accuracy and interpretability

Effective separation of causal and non-causal factors

Validated through comprehensive ablation studies

Abstract

Multimodal MRI offers complementary multi-scale information to characterize the brain structure. However, it remains challenging to effectively integrate multimodal MRI while achieving neuroscience interpretability. Here we propose to use Laplacian harmonics and spectral graph theory for multimodal alignment and multiscale integration. Based on the cortical mesh and connectome matrix that offer multi-scale representations, we devise Laplacian operators and spectral graph attentions to construct a shared latent space for model alignment. Next, we employ a disentangled learning combined with Graph Variational Autoencoder architectures to separate scale-specific and shared features. Lastly, we design a mutual information-informed bilevel regularizer to separate causal and non-causal factors based on the disentangled features, achieving robust model performance with enhanced interpretability. Our model outperforms baselines and other state-of-the-art models. The ablation studies confirmed the effectiveness of the proposed modules. Our model promises to offer a robust and interpretable framework for multi-scale brain structure analysis.