Non-invasive Neural Decoding in Source Reconstructed Brain Space

TL;DR

This paper introduces a method for non-invasive neural decoding that reconstructs brain activity in source space, enabling improved generalization, interpretability, and data harmonization across different neuroimaging datasets.

Contribution

It demonstrates that source reconstruction allows MEG data to be decoded in brain space, facilitating spatial biases and zero-shot generalization.

Findings

Enables spatial inductive biases in neural decoding.

Improves interpretability of MEG data.

Achieves zero-shot generalization across datasets.

Abstract

Non-invasive brainwave decoding is usually done using Magneto/Electroencephalography (MEG/EEG) sensor measurements as inputs. This makes combining datasets and building models with inductive biases difficult as most datasets use different scanners and the sensor arrays have a nonintuitive spatial structure. In contrast, fMRI scans are acquired directly in brain space, a voxel grid with a typical structured input representation. By using established techniques to reconstruct the sensors' sources' neural activity it is possible to decode from voxels for MEG data as well. We show that this enables spatial inductive biases, spatial data augmentations, better interpretability, zero-shot generalisation between datasets, and data harmonisation.