Modeling Shared Responses in Neuroimaging Studies through MultiView ICA

TL;DR

This paper introduces a novel MultiView ICA model for group neuroimaging studies that effectively captures shared brain sources across subjects, demonstrating improved sensitivity, lower variability, and more accurate source localization in fMRI and MEG data.

Contribution

The paper presents a new MultiView ICA model with a closed-form likelihood and a robust optimization method, enhancing group neuroimaging analysis for complex, naturalistic stimuli.

Findings

Improved sensitivity in identifying common sources among subjects.

Lower between-session variability of recovered sources.

More accurate source localization in MEG data.

Abstract

Group studies involving large cohorts of subjects are important to draw general conclusions about brain functional organization. However, the aggregation of data coming from multiple subjects is challenging, since it requires accounting for large variability in anatomy, functional topography and stimulus response across individuals. Data modeling is especially hard for ecologically relevant conditions such as movie watching, where the experimental setup does not imply well-defined cognitive operations. We propose a novel MultiView Independent Component Analysis (ICA) model for group studies, where data from each subject are modeled as a linear combination of shared independent sources plus noise. Contrary to most group-ICA procedures, the likelihood of the model is available in closed form. We develop an alternate quasi-Newton method for maximizing the likelihood, which is robust and converges quickly. We demonstrate the usefulness of our approach first on fMRI data, where our model demonstrates improved sensitivity in identifying common sources among subjects. Moreover, the sources recovered by our model exhibit lower between-session variability than other methods.On magnetoencephalography (MEG) data, our method yields more accurate source localization on phantom data. Applied on 200 subjects from the Cam-CAN dataset it reveals a clear sequence of evoked activity in sensor and source space. The code is freely available at https://github.com/hugorichard/multiviewica.