Selective detrending method for reducing task‐correlated motion artifact during speech in event‐related FMRI

Abstract

Task-correlated motion artifacts that occur during functional magnetic resonance imaging can be mistaken for brain activity. In this work, a new selective detrending method for reduction of artifacts associated with task-correlated motion (TCM) during speech in event-related functional magnetic resonance imaging is introduced and demonstrated in an overt word generation paradigm. The performance of this new method is compared with that of three existing methods for reducing artifacts because of TCM: (1) motion parameter regression, (2) ignoring images during speech, and (3) detrending time course datasets of signal components related to TCM (deduced from artifact corrupted voxels). The selective detrending method outperforms the other three methods in reducing TCM artifacts and in retaining blood oxygenation level dependent signal.

Regional changes in word-production laterality after a naming treatment designed to produce a rightward shift in frontal activity

“…In brief, this procedure (1) estimates residual-motion-related artifacts by selecting deconvolved responses (DRs) from several voxels outside the brain with high R 2 values, and then (2) removes the artifact most highly correlated with each brain voxel’s time series when this correlation exceeds .50. (See Crosson et al (2007a), Gopinath (2003), and Gopinath et al (2009) for details.) For one patient’s (02-036’s) pre-treatment images, task-correlated artifacts had durations too long (15 – 20 sec) to be reduced by the selective detrending algorithm, so a second detrending technique was employed in addition to selective detrending.…”

An fMRI Study of the Differences in Brain Activity During Active Ankle Dorsiflexion and Plantarflexion

“…To correct for this, the deconvolved data from each participant was visually inspected to determine whether motion artifact occurred coincident with the onset of the ankle movement. Based on this inspection, stimulus-correlated motion was seen in four participants and a detrending procedure was employed to remove this motion artifact (Gopinath et al 2008). Functional images were spatially smoothed using a 3 mm Gaussian kernel full-width at half-maximum.…”

Neuroimaging in aphasia treatment research: Consensus and practical guidelines for data analysis

“…However, even when using such designs, overt articulation may still result in false positive activity and standard detrending algorithms that aim to remove motion-related signal from the time series non-selectively across all voxels, may result in reduced sensitivity (see Crosson et al, 2007 for details). More recently developed detrending algorithms consider the latter weakness by selectively removing motion related signal changes from the images, which results in improved sensitivity and specificity (Gopinath et al, 2009). …”

“…The disadvantage to this approach is that it is very sensitive to noise such as that generated during overt speech during scanning. In the latter instance it is necessary to have a technique for minimizing noise in the data (e.g., Gopinath et al, 2009), and such techniques are not perfect. Averaging of raw signals from response epochs timed to experimental manipulations is also assumption free, but is subject to distortions, for instance when there are sequential dependencies of HRFs (Serences, 2004).…”

“…Data also can be extracted from two sessions separately for subsequent comparison, and then some procedure should be used to ensure that the detection sensitivity for activation is equivalent across sessions (Parrish et al, 2000). For example, Gopinath et al (2009) developed a technique to compensate for differences in detection sensitivity of BOLD measures across sessions. In short, the technique starts with the residuals of the regression of a deconvolved hemodynamic response series against the acquired time series, and uses this as the starting point for a mixed auto-regressive plus white noise model to estimate noise structure for two sessions on a voxel-by-voxel basis.…”