Short TE FMRI Data: Removing Motion and Physiological Noise Confounds from BOLD FMRI

Molly Gallogly Bright¹, Kevin Murphy¹

FMRI data reflect noise sources in addition to neuronal activation, and their relative contributions depend on the echo time (TE). Short TE data may contain information related to motion and physiology, and could potentially correct BOLD-weighted fMRI data for these confounds. We used a dual-echo sequence to simultaneously record fMRI data at short and BOLD-weighted TEs, introducing head nodding, breathing challenges, and visual stimulation to characterize their effects on both datasets. Significant correlations were observed between short TE data and noise regressors. Using short TE data to correct BOLD-weighted data (replacing motion correction regressors) increased the significance of activation maps.