Decoding the individual finger movements from single‐trial functional magnetic resonance imaging recordings of human brain activity

• DOI: 10.1111/ejn.12547
• Publication Date: 2014-03-24T11:30:50Z
• Authors: Guohua Shen; Jing Zhang; Mengxing Wang; Du Lei; Guang Yang; Shanmin Zhang; Xiaoxia Du
• Keywords: Finger tapping

Abstract Multivariate pattern classification analysis ( MVPA ) has been applied to functional magnetic resonance imaging (f MRI data decode brain states from spatially distributed activation patterns. Decoding upper limb movements non‐invasively recorded human is crucial for implementing a brain–machine interface that directly harnesses an individual's thoughts control external devices or computers. The aim of this study was the individual finger f single‐trial data. Thirteen healthy subjects participated in visually cued delayed movement task, and only one slight button press performed each trial. Using , decoding accuracy DA computed separately different motor‐related regions interest. For construction feature vectors, vectors two successive volumes image series trial were concatenated. With these spatial–temporal we obtained 63.1% average (84.7% best subject) contralateral primary somatosensory cortex 46.0% (71.0% motor cortex; both values significantly above chance level (20%). In addition, implemented searchlight search informative unbiased manner across whole brain. Furthermore, by applying volume trial, demonstrated information decoding, temporally. results suggest non‐invasive technique may provide features potential developing ‐based movement.