Multimodal subspace independent vector analysis effectively captures the latent relationships between brain structure and function

Abstract

A key challenge in neuroscience is to infer the relationships between brain structure and function from high-dimensional, multimodal neuroimaging data. While conventional multivariate approaches often simplify statistical assumptions and estimate one-dimensional independent sources shared across modalities, the true relationships between latent sources are likely more complex--statistical dependence may exist both within and between modalities, and possibly span more than one dimension per modality. Here we present Multimodal Subspace Independent Vector Analysis (MSIVA), a methodology to capture both joint and unique vector sources from multiple data modalities by defining both cross-modal and unimodal subspaces with variable dimensions. In particular, MSIVA enables flexible estimation of varying-size independent subspaces within modalities and their one-to-one linkage to corresponding subspaces across modalities. A key advantage is its ability to capture subjectlevel variability at the voxel level within independent subspaces, in contrast to traditional methods that share the same independent components across subjects. We evaluated three initialization workflows with five candidate subspace structures in multiple synthetic datasets and two large multimodal neuroimaging datasets, both including structural MRI (sMRI) and functional MRI (fMRI). After confirming that MSIVA successfully recovered the ground-truth subspace structures in the synthetic data, we then applied MSIVA to identify the latent subspace structure in the neuroimaging data. Subsequent subspace-specific canonical correlation analysis, brain-phenotype prediction, and voxelwise brain-age delta analysis suggest that the estimated sources from MSIVA with the optimal subspace structure are strongly associated with multiple phenotype variables, including age, sex, schizophrenia, lifestyle factors, and cognitive functions. Further, we identified modality- and group-specific brain regions related to multiple phenotype measures such as age (for example, cerebellum, precentral gyrus, and cingulate gyrus in sMRI; occipital lobe and superior frontal gyrus in fMRI), sex (for example, cerebellum in sMRI, frontal lobe in fMRI, and precuneus in both sMRI and fMRI), schizophrenia (for example, cerebellar, frontal, and insular cortices in sMRI; occipital pole, lingual gyrus, and precuneus in fMRI), shedding light on phenotypic and neuropsychiatric biomarkers of linked brain structure and function.