Fig. 5: Comparing SAXS-TT outcomes to histology and MRI.
a Voxel-wise Pearson (r) and Spearman (ρ) correlations against histology- and SAXS-TT-derived myelin levels. All p values are 0, correlations over n = 361,042 voxels. Figures below table depict the same sagittal slice of the compared parameter maps. MTsat magnetization transfer saturation, MTR magnetization transfer ratio, FA fractional anisotropy, RD radial diffusivity, RK radial kurtosis, AWF axonal water fraction, DTI diffusion tensor imaging, DKI diffusion kurtosis imaging. b First against second row of table in a, identity line (dotted), linear fit (black line) with indicated equation, and a Pearson correlation of r = 0.996 (p = 1.1 × 10−7). c Third against fourth row of table in a, identity line (dotted), linear fit (black line) with indicated equation, and a Pearson correlation of r = 0.998 (p = 2.5 × 10−8). d, e, g, h Coronal mouse brain (d, e) and horizontal mouse spinal cord (g, h) sections depicting SAXS-TT- and dMRI-derived fiber orientations, intensity weighted by SAXS-TT myelin levels, orientations interpreted by red-green-blue arrows. cc corpus callosum, cpd cerebral peduncle, ci cingulum, df dorsal fornix, em external medullary lamina of thalamus, ph posterior horn. f, i Dot product between SAXS-TT- and dMRI-derived fiber orientations for brain (f) and spinal cord (i), where dot-product value of 1 denotes perfect collinearity, and 0 a 90° discrepancy. Source data are provided as a Source Data file.
