Noise in Functional Magnetic Resonance Imaging

摘要

Functional magnetic resonance imaging (fMRI) is a method for analyses and to capture dynamic patterns of process occurring in the brain of a human being. The fMRI measurement is based on the tissue parameters and physics behind MRI. fMRI produce highly dimensional, sparse and noise in data which are crucial to envision, analyse and monitor. The method of smoothing by spatial domain is frequently used in fMRI studies because fMRI data represent the correlations in spatial domain due to similarities of brain regions and soften of the vascular system. In this paper, time - series noise in fMRI is pre-processed by filters and their SNR are compared. Different parameters of noise have been studied.