The human brain is a complex dynamic system. Understanding the functional difference of the brain between the resting state and performing tasks helps to decipher the functional brain architecture. In this work, we trained an autoencoder to optimally reconstruct the resting state fMRI (rs-fMRI) data of 50 subjects, and this model was subsequently used to reconstruct the movie watching fMRI (mw-fMRI) data of the same subjects. We have compared the reconstruction error between movie watching and resting state, and significant regional functional differences were found.
This abstract and the presentation materials are available to members only; a login is required.