Utility of real-time prospective motion correction (PROMO) on 3D T1-weighted imaging in automated brain structure measurements

Abstract

PROspective MOtion correction (PROMO) can prevent motion artefacts. The aim of this study was to determine whether brain structure measurements of motion-corrected images with PROMO were reliable and equivalent to conventional images without motion artefacts. The following T1-weighted images were obtained in healthy subjects: (A) resting scans with and without PROMO and (B) two types of motion scans (“side-to-side” and “nodding” motions) with and without PROMO. The total gray matter volumes and cortical thicknesses were significantly decreased in motion scans without PROMO as compared to the resting scans without PROMO (p < 0.05). Conversely, Bland–Altman analysis indicated no bias between motion scans with PROMO, which have good image quality, and resting scans without PROMO. In addition, there was no bias between resting scans with and without PROMO. The use of PROMO facilitated more reliable brain structure measurements in subjects moving during data acquisition.

Effectiveness of navigator-based prospective motion correction in MPRAGE data acquired at 3T

“…Subject motion produced a consistent bias of reduced cortical thickness and volume, as extracted from both the surface-based and default streams of FreeSurfer. These results are consistent with those of previously published works [ 11 , 16 ]. Although there was some improvement in accuracy of cortical measures when either the FOV-update or Reacquisition were used, utilizing both PMC components provided even greater accuracy, resulting in no significant differences compared to a non-moving subject.…”

“…PROMO is ideal for studies of pediatric or adult patient populations as no additional hardware needs to be placed on the subject [ 13 ]. When utilized within an MPRAGE sequence, PROMO has been previously shown to reduce errors in cortical surface reconstructions [ 12 , 14 ], as well as remove bias in total gray matter volume and cortical thickness estimates for moving subjects [ 15 , 16 ]. Navigator-based PMC techniques can employ two strategies to improve image quality: 1) updating the gradients and RF pulses to ensure excitation and acquisition of the same brain slice/slab and imaging field-of-view (FOV) in every TR (hereafter referred to as ‘FOV-update’), and 2) reacquisition of k-space data in which excessive motion has occurred between navigators acquired immediately before and after a particular k-space segment (hereafter referred to as ‘Reacquisition’).…”

Improved cortical surface reconstruction using sub-millimeter resolution MPRAGE by image denoising

“…This versatility is beneficial for obtaining high-quality images at a true, high sub-millimeter resolution using only product sequences and reconstruction provided by vendors, which can only encode a whole-brain volume at a high resolution nominally because of the use of partial Fourier imaging and the increased T1 blurring during the inversion recovery. Simultaneous denoising and super-resolution is also helpful for obtaining images at ultra-high resolution (e.g., 0.25-mm isotropic resolution 126 , which requires averaging 8 repetitions from 10 hour scan on a 7-Tesla scanner) by making a balance between extremely challenging individual denoising or super-resolution and avoiding the use of prospective motion correction to acquire input noisy images at native ultra-high resolution for denoising [127][128][129][130][131] . Image super-resolution has also been shown to be an alternative way to obtain high-quality sub-millimeter resolution images for improving cortical surface reconstruction 47 .…”

Method for retrospective estimation of natural head movement during structural MRI

“…These artifacts may be particularly problematic when using structural T 1 ‐weighted MRI for morphometric analyses, and particularly when trying to understand morphometric changes in clinical populations. Recent studies, where participants were instructed to perform head movement during structural MRI acquisition, show that head motion significantly biases brain morphometry estimates and test–retest reliability . There is, therefore, an interest in identifying head motion and characterizing its potential effects such that morphometry biases can be minimized.…”

“…Recent studies, where participants were instructed to perform head movement during structural MRI acquisition, show that head motion significantly biases brain morphometry estimates and test-retest reliability. [5][6][7] There is, therefore, an interest in identifying head motion and characterizing its potential effects such that morphometry biases can be minimized.…”