Limits...
An evaluation of prospective motion correction (PMC) for high resolution quantitative MRI.

Callaghan MF, Josephs O, Herbst M, Zaitsev M, Todd N, Weiskopf N - Front Neurosci (2015)

Bottom Line: In the presence of head motion, PMC-based motion correction considerably improved the quality of the maps as reflected by fewer visible artifacts and improved consistency.The precision of the maps, parameterized through the coefficient of variation in cortical sub-regions, showed improvements of 11-25% in the presence of deliberate head motion.Such a robust motion correction scheme is crucial in order to achieve the ultra-high resolution required of quantitative imaging for cutting edge in vivo histology applications.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Centre for Neuroimaging, UCL Institute of Neurology, University College London London, UK.

ABSTRACT
Quantitative imaging aims to provide in vivo neuroimaging biomarkers with high research and diagnostic value that are sensitive to underlying tissue microstructure. In order to use these data to examine intra-cortical differences or to define boundaries between different myelo-architectural areas, high resolution data are required. The quality of such measurements is degraded in the presence of motion hindering insight into brain microstructure. Correction schemes are therefore vital for high resolution, whole brain coverage approaches that have long acquisition times and greater sensitivity to motion. Here we evaluate the use of prospective motion correction (PMC) via an optical tracking system to counter intra-scan motion in a high resolution (800 μm isotropic) multi-parameter mapping (MPM) protocol. Data were acquired on six volunteers using a 2 × 2 factorial design permuting the following conditions: PMC on/off and motion/no motion. In the presence of head motion, PMC-based motion correction considerably improved the quality of the maps as reflected by fewer visible artifacts and improved consistency. The precision of the maps, parameterized through the coefficient of variation in cortical sub-regions, showed improvements of 11-25% in the presence of deliberate head motion. Importantly, in the absence of motion the PMC system did not introduce extraneous artifacts into the quantitative maps. The PMC system based on optical tracking offers a robust approach to minimizing motion artifacts in quantitative anatomical imaging without extending scan times. Such a robust motion correction scheme is crucial in order to achieve the ultra-high resolution required of quantitative imaging for cutting edge in vivo histology applications.

No MeSH data available.


Related in: MedlinePlus

Spatial maps of the coefficient of variation of R1 for three volunteers. Motion increased the CoV, but to a lesser extent when using the PMC system. In the absence of intentional motion, the CoV was increased by the PMC system for volunteer 1 (top row) but reduced for all other volunteers, e.g., volunteer 3 (middle row) and volunteer 4. Volunteer 4 (bottom row) moved very rapidly in the motion condition limiting the improvement gained by the PMC system. The rapid motion resulted in poor segmentation for the motion, PMC off condition.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4373264&req=5

Figure 3: Spatial maps of the coefficient of variation of R1 for three volunteers. Motion increased the CoV, but to a lesser extent when using the PMC system. In the absence of intentional motion, the CoV was increased by the PMC system for volunteer 1 (top row) but reduced for all other volunteers, e.g., volunteer 3 (middle row) and volunteer 4. Volunteer 4 (bottom row) moved very rapidly in the motion condition limiting the improvement gained by the PMC system. The rapid motion resulted in poor segmentation for the motion, PMC off condition.

Mentions: Figure 3 shows spatial maps of the CoV of R1 from three volunteers in AAL-defined regions of interest. Motion greatly increased the CoV (columns three and four), but to a lesser extent when the PMC system was on (column four). The impact of the motion and PMC factors varied spatially. Within the condition of motion, PMC off the CoV was higher anteriorly (column three) reflecting the fact that movement of the head was restricted posteriorly with volunteers in the supine position. Within the condition of motion, PMC on the CoV was higher inferiorly (column 4, e.g., in the cerebellum) where the assumption of rigid body motion is less valid (Greitz et al., 1992; Soellinger et al., 2009). Figure 4 summarizes the group differences in the global CoV measure relative to the no motion, PMC off condition for each map. In the motion, PMC off case the CoV was significantly increased (Figure 4B, p = 0.0156 for each map). The increases were 0.042 ± 0.012 (median ± inter-quartile range) for PD*, 0.039 ± 0.012 for MT, 0.076 ± 0.039 for R1 and 0.125 ± 0.039 for R*2. These changes corresponded to a median CoV increase with respect to the no motion, PMC off condition of 52.6, 21.3, 52.7, and 39.1% for PD*, MT, R1 and R*2, respectively (Figure 4B) showing that motion had the greatest impact on variance levels in the PD* and R1 maps.


An evaluation of prospective motion correction (PMC) for high resolution quantitative MRI.

Callaghan MF, Josephs O, Herbst M, Zaitsev M, Todd N, Weiskopf N - Front Neurosci (2015)

Spatial maps of the coefficient of variation of R1 for three volunteers. Motion increased the CoV, but to a lesser extent when using the PMC system. In the absence of intentional motion, the CoV was increased by the PMC system for volunteer 1 (top row) but reduced for all other volunteers, e.g., volunteer 3 (middle row) and volunteer 4. Volunteer 4 (bottom row) moved very rapidly in the motion condition limiting the improvement gained by the PMC system. The rapid motion resulted in poor segmentation for the motion, PMC off condition.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4373264&req=5

Figure 3: Spatial maps of the coefficient of variation of R1 for three volunteers. Motion increased the CoV, but to a lesser extent when using the PMC system. In the absence of intentional motion, the CoV was increased by the PMC system for volunteer 1 (top row) but reduced for all other volunteers, e.g., volunteer 3 (middle row) and volunteer 4. Volunteer 4 (bottom row) moved very rapidly in the motion condition limiting the improvement gained by the PMC system. The rapid motion resulted in poor segmentation for the motion, PMC off condition.
Mentions: Figure 3 shows spatial maps of the CoV of R1 from three volunteers in AAL-defined regions of interest. Motion greatly increased the CoV (columns three and four), but to a lesser extent when the PMC system was on (column four). The impact of the motion and PMC factors varied spatially. Within the condition of motion, PMC off the CoV was higher anteriorly (column three) reflecting the fact that movement of the head was restricted posteriorly with volunteers in the supine position. Within the condition of motion, PMC on the CoV was higher inferiorly (column 4, e.g., in the cerebellum) where the assumption of rigid body motion is less valid (Greitz et al., 1992; Soellinger et al., 2009). Figure 4 summarizes the group differences in the global CoV measure relative to the no motion, PMC off condition for each map. In the motion, PMC off case the CoV was significantly increased (Figure 4B, p = 0.0156 for each map). The increases were 0.042 ± 0.012 (median ± inter-quartile range) for PD*, 0.039 ± 0.012 for MT, 0.076 ± 0.039 for R1 and 0.125 ± 0.039 for R*2. These changes corresponded to a median CoV increase with respect to the no motion, PMC off condition of 52.6, 21.3, 52.7, and 39.1% for PD*, MT, R1 and R*2, respectively (Figure 4B) showing that motion had the greatest impact on variance levels in the PD* and R1 maps.

Bottom Line: In the presence of head motion, PMC-based motion correction considerably improved the quality of the maps as reflected by fewer visible artifacts and improved consistency.The precision of the maps, parameterized through the coefficient of variation in cortical sub-regions, showed improvements of 11-25% in the presence of deliberate head motion.Such a robust motion correction scheme is crucial in order to achieve the ultra-high resolution required of quantitative imaging for cutting edge in vivo histology applications.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Centre for Neuroimaging, UCL Institute of Neurology, University College London London, UK.

ABSTRACT
Quantitative imaging aims to provide in vivo neuroimaging biomarkers with high research and diagnostic value that are sensitive to underlying tissue microstructure. In order to use these data to examine intra-cortical differences or to define boundaries between different myelo-architectural areas, high resolution data are required. The quality of such measurements is degraded in the presence of motion hindering insight into brain microstructure. Correction schemes are therefore vital for high resolution, whole brain coverage approaches that have long acquisition times and greater sensitivity to motion. Here we evaluate the use of prospective motion correction (PMC) via an optical tracking system to counter intra-scan motion in a high resolution (800 μm isotropic) multi-parameter mapping (MPM) protocol. Data were acquired on six volunteers using a 2 × 2 factorial design permuting the following conditions: PMC on/off and motion/no motion. In the presence of head motion, PMC-based motion correction considerably improved the quality of the maps as reflected by fewer visible artifacts and improved consistency. The precision of the maps, parameterized through the coefficient of variation in cortical sub-regions, showed improvements of 11-25% in the presence of deliberate head motion. Importantly, in the absence of motion the PMC system did not introduce extraneous artifacts into the quantitative maps. The PMC system based on optical tracking offers a robust approach to minimizing motion artifacts in quantitative anatomical imaging without extending scan times. Such a robust motion correction scheme is crucial in order to achieve the ultra-high resolution required of quantitative imaging for cutting edge in vivo histology applications.

No MeSH data available.


Related in: MedlinePlus