This paper presents a novel MVPA method based on deep neural networks, which can identify a group of voxels with their pattern of activity capable of differentiating experimental conditions. Through the forward inference procedure, the proposed deep neural network can also be applied to distinguish brain imaging data of different experimental conditions. Our experimental results suggest that deep neural networks are of great potential as an MVPA tool for functional brain mapping.
This abstract and the presentation materials are available to members only; a login is required.