Detection of self-paced movement intention from pre-movement electroencephalogram signals with Hilbert transform

The movement related cortical potential (MRCP) is a well-known neural signature of human's self-paced movement intention, which can be exploited by future neuroprosthesis. Most existing studies have explored the amplitude representation for the movement intention. In this paper we investigate the Hilbert transformed MRCP, which implicitly includes phase information complementarily to amplitude information, for the detection of self-paced upper-limb movement intention. On the datasets in which 5 healthy subjects executed a self-initiated upper limb center-out reaching task in three sessions, we have evaluated the detection model with Hilbert transformed MRCP as features and the state-of-art one with the original MRCP amplitude as features. Results show that the Hilbert transformed MRCP based detector is more accurate than the original MRCP based one.