Table 3.
MR imaging specifications across studies.
| Scan type | Scanner model and head coil | Diffusion settings and directions | MRI scans settings | Analysis type | Analysis specifications | |
|---|---|---|---|---|---|---|
| Jang et al. (2020b) | DWI | 1.5T Phillips Gyroscan Intera6 channel head coil | b = 1,000 s/mm2 | Acquisition matrix = 96 × 96; reconstructed matrix = 192 × 192; FOV = 240 × 240 mm; TR = 10,398 ms; TE = 72 ms; EPI factor = 59; slice thickness=2.5 mm | Seed based Tractography FA; TV; Fazekas Grade | FMRIB Diffusion Software with routines option (0.5 mm step lengths, 5,000 streamline samples, curvature threshold = 0.2) used for fiber tracking |
| Jang et al. (2020a) | DWI | 1.5T Phillips Gyroscan Intera6 channel head coil | b = 1,000 s/mm2 | Acquisition matrix=96 × 96; reconstructed to matrix = 192 × 192; FOV = 240 × 240 mm2; TR = 10,398 ms; TE = 72 ms; SENSE factor = 2; EPI factor = 59; slice thickness = 2.5 mm. | Seed based Tractography FA; TV; Modified Graeb Score | FMRIB Diffusion Software with routines option(0.5 mm step lengths, 5,000 streamline samples, curvaturethresholds = 0.2) used for fiber tracking; analysis done by expert w/3 years of experience |
| Lee et al. (2020) | DWI, FLAIR | NR | NR | NR | Fazekas grade (manual rating) | Two physiatrists completed analysis |
| Fandler et al. (2017) | T2- axial fast spin echo, axial T2 FLAIR, sagittal T1 spin echo, gradient echo T2, axial diffusion-weighted single shot echo planar axial | 1.5T Siemens Symphony | NR | All axial slices slice thickness = 5 mm | Manual lesion identification; Fazekas Grade | Two independent experts used standards for reporting vascular changes on neuroimaging consensus criteria |
| Fandler et al. (2018) | T2- axial fast spin echo sequence, FLAIR sequence–axial, gradient echo T2, DWI single-shot echo planar with ADC maps (axial) and TOP angiography | 1.5 T | NR | Axial T2-weighted fast spin echo sequence (0.5 × 0.5 × 5 mm); FLAIR sequence (0.4 × 0.4 × 5 mm); sagittal T1-weighted spin echo sequence (0.6 × 0.6 × 5 mm); gradient echo T2* weighted sequence (0.4 × 0.4 × 5 mm); axial DWI single-shot echo planar imaging sequence (1.2 × 1.2 × 5 mm) with apparent diffusion coefficient ADC maps and a 3D time of flight (TOF) angiography. Axial slices; slice thickness = 5 mm with 0.5 mm gap | Lesion probability mapping; tractography analysis | RSSI manually marked by two neuroimaging experts |
| Moon et al. (2017) | T2 FLAIR and DWI | NR | NR | NR | Fazekas scale (using diffusion images with FLAIR MRI); unspecified localization | Physician rating |
| Li et al. (2014) | Resting state fMRI, DTI | 3T MRI GE Signa EXCITE | b = 1,000 s/mm2 reference scan with b = 0 (no diffusion gradient) 15 non-collinear directions | NR | Seed-based functional connectivity maps (from primary motor and supplementary motor to the brain); swallowing-related functional connectivity for 20 ROIs; mean FA between the SMA and M1 | Whole brain fiber tracking using “Diffusion Toolkit” software |
| Mihai et al. (2016) | DWI, T1-weighted, fMRI | 3T Siemens Verio, 32-channel head coil | b = 1,000 s/mm2; 64 gradient directions | Gradient echo (34 slices 2 × 2 × 2 mm); functional EPI (96 × 96 oblique); structural T1 (1 × 1 × 1) | FA lateralization index; T1 lesion size | FSL, MNI, and bedpostx were used for probabilistic tractography; researchers manually drew borders for lesions for lesion size, for detailed description of other programs used see paper. |
| Galovic et al. (2017) | Transverse T2, T1, FLAIR, saggital T2, isotropic DWI | I.5T Siemens Avanto, 1.5T Siemens Symphony, or 3T Siemens Verio MRI | DWI (b = 1,000 s/mm2 with 4 mm transverse slices) | FLAIR slice thickness = 5 mm; T2 sagittal slice thickness = 4.5 mm; transverse DWI slice thickness = 4 mm | Voxel-based lesion symptom mapping (VLSM); ROI analysis; Probabilistic Tractography for healthy individuals | VLSM: ICBM standard brain template, NPM software.Tractography: FMRIB Diffusion toolkit, FSL FLIRT algorithm |
| Flowers et al. (2017) | Saggital T1, T2 FLAIR, isotropic axial diffusion | 1.5 T Signa EchoSpeech MR scanner (GE), quadrature head coil | b = 1,000 s/mm2 | T1: 7.5 mm slice thickness 2 mm space; T2: 5 mm slice thickness = 2 mm spacing; diffusion slice thickness = 5 mm slice w/0 mm spacing | Fazekas scale for periventricular hyperintensities and deep hyperintensities in 12 ROIs | Manually traced lesions on each DWI slice with MRIcron and calculated volumes |
| Galovic et al. (2013) | T2, T1, FLAIR, Saggital T2, isotropic DWI, with TOF sequence | 1.5T Siemens Avanto 1.5T Siemens Symphony or 3T Siemens Verio | b = 1,000 s/m2 | FLAIR slice thickness = 5 mm; sagittal T2 slice thickness = 4.5 mm; DWI slice thickness = 4 mm | Lesion mapping with ROI | MICRON with MNI space for model. Semi-automatic image analysis by 1 neurologist using MIPAV to Talairach (used Broadmann areas) and visually noted lesions with binomial scale. ImageJ used for lesion size. Age-related white matter documented by Wahlund et al. (2001) |
| Wilmskoetter et al. (2019) | DWI | NR | NR | Voxel-wise resolution ranged from 0.9375 × 0.9375 × 3 mm to 1.4458 × 1.4458 × 6 mm | Manually drawn lesions; VLSM; ROI when there were no significant findings from VLSM | Lesions drawn by a researcher using MRIcron; reviewed by neurologist with expertise in VLSM; custom MATLAB script for lesion symptom mapping |
| Levine et al. (1992) | T2 | 1.5T GE Signa | n/a | NR | T2 scales graded according to Awad et al. (1986); number of unidentified bright objects | NR |
| Kumar et al. (2012) | DWI | NR | NR | NR | Lesion volume and location on DWI | Image J; brain atlas |
| Jang et al. (2017) | DTI (at both 5 and 9 weeks) | 1.5T Philips Gyroscan Intera | 1,000 s/mm2 | 60 continuous slices; 76 ms; 2.5 mm thickness for each of the 32 gradients | Probabilistic tractography; ROI | FSL FMRIB |
| Cola et al. (2010) | DWI | NR | b = 1,000 s/mm2; 3 directions | Slice thickness = 6 mm axial images; | Lesion volumes semi-manually tracked | Image J (semiautomatic threshold) |
| Ko et al. (2019) | NR | NR | NR | Slice thickness = 3-5 mm | Lesion size using Sims et al. (2009) method; Fazekas scale | Expert |
| Mouräo et al. (2017) | T1 NPRAGE, MRI, DTI | 3T Siemens Magnetom Trio; 32-channel head coil | b = 800 s/mm2; 64 directions | T1 and DTI: slice thickness = 2 mm, 75 slices | Tractography; fractional anisotropy; radial diffusivity; mean diffusivity; fibers count | DTI studio |
| Suntrup et al. (2015) | DWI, T2 FLAIR, | MRI: 1.5T Intera Gyroscan, Philips. CT: somatom Definition AS+ Siemens | b = 1,000 s/m2** | MRI: slice thickness = 5 mm transverse. reconstruction, 1 mm increment | Atlas-based regional analysis (% of brain affected) | Completed by neuroradiologist using FSL FLIRT, FMRIB and FNIRT |
| Galovic et al. (2016) | T2 (transverse and sagital), T1, FLAIR, and DWI | 1.5T Siemens Avanto, 1.5T Siemens Symphony or 3T Siemens Verio | b = 1,000 s/m2 | DWI transverse slices 4 mm; T2: T1: FLAIR slice thickness = 5 mm; Sagittal T2: slice thickness = 4.5 mm | Voxel-based lesion symptom mapping (VLSM) | VLSM was calculated using MRIcron software |
| Wan et al. (2016)* | MRI (no details given) | NR | NR | NR | NR, Though presence of stroke was determined by scans | NR |
| Kim et al. (2014)* | MRI (no details given) | NR | NR | NR | Determined vascular territory with modified Rovira et al. (2005) method | NR |
*
Received a modified NIH Quality Assessment rating of “poor”.
**
The authors reported this as “b = 0 and 1.9 mm/s2” which we standardized to b = 1,000 s/m2.
Order of appearance: DWI, diffusion weighted imaging; FA, fractional anisotropy; TV, tract volume; FMRIB, Functional Magnetic Resonance Imaging of the Brain; FLAIR, fluid attenuated inversion recovery; NR, not rated; RSSI, MRI magnetic resonance imaging; fMRI, functional magnetic resonance imaging; DTI, diffusion tensor imaging; SM, Supplementary motor area and M1; EPI, Echo-planar imaging; FSL, FMRIB Software Library; MNI, Montreal Neurological Institute; VLSM, voxel-based lesion symptom mapping; ICBM, International Consortium for Brain Mapping; NPM, FLIRT FMRIB's Linear Image Registration Tool; MR, Magnetic resonance; MPRAGE, magnetization-prepared rapid gradient-echo; ADC, Apparent Diffusion Coefficient; FACT, Fiber Assignment by Continuous Tracking, apparent diffusion coefficient A.