Impact of image acquisition on voxel-based-morphometry investigations of age-related structural brain changes

• Published in: NeuroImage (Orlando, Fla.) Vol. 87; pp. 170 - 182
• Main Authors: Streitbürger, Daniel-Paolo, Pampel, André, Krueger, Gunnar, Lepsien, Jöran, Schroeter, Matthias L., Mueller, Karsten, Möller, Harald E.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 15.02.2014; Elsevier; Elsevier Limited
• Subjects: Acquisitions; Adult; Aged; Aging; Aging - pathology; Arrays; Bias; Biological and medical sciences; Brain; Brain - pathology; Classification; Coils; Female; Fourier transforms; Fundamental and applied biological sciences. Psychology; Geriatrics; Humans; Image acquisition; Image Interpretation, Computer-Assisted - methods; Imaging pulse sequence; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Marketing; Middle Aged; Morphometry; Neuroimaging; Noise; Radiofrequency coil; Standard deviation; Strategy; Studies; Vertebrates: nervous system and sense organs; Voxel size; Voxel-based morphometry; Young Adult; Aging; Imaging pulse sequence; Radiofrequency coil; Voxel size; Voxel-based morphometry; Senescence; Imaging; Central nervous system; Morphometry; Age; Encephalon
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2013.10.051

A growing number of magnetic resonance imaging studies employ voxel-based morphometry (VBM) to assess structural brain changes. Recent reports have shown that image acquisition parameters may influence VBM results. For systematic evaluation, gray-matter-density (GMD) changes associated with aging were investigated by VBM employing acquisitions with different radiofrequency head coils (12-channel matrix coil vs. 32-channel array), different pulse sequences (MP-RAGE vs. MP2RAGE), and different voxel dimensions (1mm vs. 0.8mm). Thirty-six healthy subjects, classified as young, middle-aged, or elderly, participated in the study. Two-sample and paired t-tests revealed significant effects of acquisition parameters (coil, pulse sequence, and resolution) on the estimated age-related GMD changes in cortical and subcortical regions. Potential advantages in tissue classification and segmentation were obtained for MP2RAGE. The 32-channel coil generally outperformed the 12-channel coil, with more benefit for MP2RAGE. Further improvement can be expected from higher resolution if the loss in SNR is accounted for. Use of inconsistent acquisition parameters in VBM analyses is likely to introduce systematic bias. Overall, acquisition and protocol changes require careful adaptations of the VBM analysis strategy before generalized conclusion can be drawn. •Investigating effects of acquisition on voxel-based morphometry detecting brain changes•Significant impact of acquisition parameters on voxel-based morphometry results•Mixing of acquisition parameters in voxel-based morphometry studies can falsify results.