Non-invasive Neural Decoding in Source Reconstructed Brain Space
This approach addresses difficulties in combining datasets and building models for brainwave decoding, which is incremental as it applies established reconstruction techniques to MEG data.
The paper tackled the problem of non-invasive brainwave decoding from MEG/EEG data by reconstructing neural activity in brain space instead of using sensor measurements, enabling spatial inductive biases, data augmentation, interpretability, zero-shot generalization between datasets, and data harmonization.
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.