Evaluation of preprocessing steps to compensate for magnetic field distortions due to body movements in BOLD fMRI

Abstract

Blood oxygenation level-dependent (BOLD) functional magnetic resonance imaging (fMRI) is currently the dominant technique for non-invasive investigation of brain functions. One of the challenges with BOLD fMRI, particularly at high fields, is compensation for the effects of spatiotemporally varying magnetic field inhomogeneities (DeltaB(0)) caused by normal subject respiration and, in some studies, movement of the subject during the scan to perform tasks related to the functional paradigm. The presence of DeltaB(0) during data acquisition distorts reconstructed images and introduces extraneous fluctuations in the fMRI time series that decrease the BOLD contrast-to-noise ratio. Optimization of the fMRI data-processing pipeline to compensate for geometric distortions is of paramount importance to ensure high quality of fMRI data. To investigate DeltaB(0) caused by subject movement, echo-planar imaging scans were collected with and without concurrent motion of a phantom arm. The phantom arm was constructed and moved by the experimenter to emulate forearm motions while subjects remained still and observed a visual stimulation paradigm. These data were then subjected to eight different combinations of preprocessing steps. The best preprocessing pipeline included navigator correction, a complex phase regressor and spatial smoothing. The synergy between navigator correction and phase regression reduced geometric distortions better than either step in isolation and preconditioned the data to make them more amenable to the benefits of spatial smoothing. The combination of these steps provided a 10% increase in t-statistics compared to only navigator correction and spatial smoothing and reduced the noise and false activations in regions where no legitimate effects would occur.

FIG. 1

(a) Phantom arm and (b) the experimental set-up to replicate the movement of the subject’s arm during a reaching/grasping paradigm.

FIG. 2

Spatially varying magnetic field inhomogeneities (6 Hz peak-to-peak) in a mid-axial slice caused by the change in position of the phantom arm between its resting and grasping positions. Frequency offsets are between 3 and 4 Hz throughout much of the occipital cortex, which are at least three times the magnitude of field offsets expected from normal subject respiration.

FIG. 3

Average t-statistic for 40 movement runs normalized with respect to NC+SS (FWHM = 7.5 mm) for each run. Although there is no significant difference between NC and NC+PR, there is a 10.1% increase between NC+PR+SS and NC+SS. Error bars represent the 95% confidence intervals (CI). [NC = navigator correction; PR = phase regression; SS = spatial smoothing; raw = neither NC nor PR]

FIG. 4

Average t-statistic in control and movement runs for the raw (unprocessed) and NC+PR+SS (navigator correction, phase regression, and spatial smoothing) pipelines. Error bars represent the 95% CI. Statistically significant differences exist between the two groups both before (p < .01) and after (p < .0001) processing.

FIG. 5

Average noise variance in VOIs through subjects’ ventricles and corpus callosum, and in signal void outside the head. Error bars represent the 95% CI. Application of PR decreased noise variance in both movement and control runs. For clarity, the vertical axis is scaled differently for each VOI. [NC = navigator correction; PR = phase regression]

FIG. 6

Activation maps for a movement run from one subject processed using (a) NC and (b) NC+PR. Green arrows show regions of false activation (Type I error) that are suppressed using the PR algorithm. (c) Raw time course for the white matter VOI outlined in (a). The timing of the visual stimulus is superimposed. The noise variance is decreased by a factor of three in the white matter VOI due to the correction of the phase discontinuity beginning around 180 seconds. [NC = navigator correction; PR = phase regression]