High resolution data analysis strategies for mesoscale human functional MRI at 7 and 9.4 T

Abstract

The advent of ultra-high field functional magnetic resonance imaging (fMRI) has greatly facilitated submillimeter resolution acquisitions (voxel volume below (1 mm³)), allowing the investigation of cortical columns and cortical depth dependent (i.e. laminar) structures in the human brain. Advanced data analysis techniques are essential to exploit the information in high resolution functional measures. In this article, we use recent, exemplary 9.4 T human functional and anatomical data to review the advantages and disadvantages of (1) pooling high resolution data across regions of interest for cortical depth profile analysis, (2) pooling across cortical depths for mapping patches of cortex while discarding depth-dependent (i.e. columnar) effects, and (3) isotropic sampling without pooling to assess individual voxel’s responses. A set of cortical depth meshes may be a solution to sampling information tangentially while keeping correspondence across depths. For quantitative analysis of the spatial organization in fine-grained structures, a cortical grid approach is advantageous. We further extend this general framework by combining it with a previously introduced cortical layer volume-preserving (equi-volume) approach. This framework can readily accommodate the research questions which allow for spatial smoothing within or across layers. We demonstrate and discuss that equi-volume sampling yields a slight advantage over equidistant sampling given the current limitations of fMRI voxel size, participant motion, coregistration and segmentation. Our 9.4 T human anatomical and functional data indicate the advantage over lower fields including 7 T and demonstrate the practical applicability of T2* and T2-weighted fMRI acquisitions.

A critical assessment of data quality and venous effects in ultra-high-resolution fMRI

“…We use isotropic voxels (0.8-mm) to ensure unbiased sampling of the convoluted cerebral 33 cortex, and we use multiband slice acceleration (Moeller et al, 2010) to achieve large coverage-such 34 coverage is important because sensory, cognitive, and motor function often reflect coordinated activity of 35 a large number of interacting brain regions. Finally, we use a modern surface-based analysis approach 36 (Glasser et al, 2013; Kemper et al, 2018;Polimeni et al, 2018), necessary for handling the convoluted 37 cortical surface visible in large field-of-view measurements (Polimeni et al, 2010). 38 39 The overarching goal in this study is to assess the quality and nature of ultra-high-resolution fMRI 40 measurements.…”

“…To this end, we devote effort to evaluating surface-based processing 42 (Sections 3.1-3.2); developing high-quality and interpretable data visualizations, especially with respect 43 to cortical folding (Section 3.3); characterizing the locations of venous effects (Sections 3.4-3.5); 44 determining whether venous effects align across subjects (Section 3.6); examining the relationship 45 between veins and BOLD responses (Section 3.7); and assessing reliability and fine-scale detail in BOLD 46 measurements (Sections 3.8-3.9). The central topic examined in this study-the problem of draining 47 veins-has been long recognized by the fMRI community (Haacke et al, 1994;Kim et al, 1994; Lai et al,…”

“…40 Abbreviations for specific sulci and gyri are as follows: IPS = intraparietal sulcus, LOS = lateral occipital 41 sulcus, TOS = transverse occipital sulcus, POS = parieto-occipital sulcus, Calc = calcarine sulcus, pSTS 42 = posterior superior temporal sulcus, IOG = inferior occipital gyrus, PLS = posterior lingual sulcus, ALS = 43 anterior lingual sulcus, OTS = occipitotemporal sulcus, CoS = collateral sulcus, FG = fusiform gyrus, 44 mFus = mid-fusiform sulcus, and ptCoS = posterior transverse collateral sulcus. 45 46 The fsaverage surface is an anatomical surface template, and FreeSurfer provides curvature-based 47 alignment of individual subjects to this template. We increased the density of the fsaverage surface (same 48 method as above) and mapped individual subjects to and from fsaverage using nearest-neighbor 49 .…”

“…42 43 In some cases, we generated a version of the data in which the final spatial interpolation is performed at 44 the positions of the 0.8-mm voxels associated with the T 1 and T 2 volumes (based on the co-registration 45 determined in the next section). Also, in some cases, to simulate low-resolution fMRI data, we spatially 46 smoothed the pre-processed functional volumes using an ideal Fourier filter (10th-order low-pass 47 . CC-BY 4.0 International license It is made available under a was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.…”

“…42 43 Bias correction of EPI intensities 44 45 EPI intensities can be used as an anatomical marker of the location of venous effects. However, MRI 46 image intensities, including those from EPI, are influenced by inhomogeneities in the transmit and receive 47 profiles of the RF coil. To reduce these biases, we first computed the mean of the surface-based pre-48 processed functional data over time, producing EPI intensities with dimensions N vertices × 6 depths.…”