Placenta microstructure and microcirculation imaging with diffusion MRI

Purpose To assess which microstructural models best explain the diffusion‐weighted MRI signal in the human placenta. Methods The placentas of nine healthy pregnant subjects were scanned with a multishell, multidirectional diffusion protocol at 3T. A range of multicompartment biophysical models were fit to the data, and ranked using the Bayesian information criterion. Results Anisotropic extensions to the intravoxel incoherent motion model, which consider the effect of coherent orientation in both microvascular structure and tissue microstructure, consistently had the lowest Bayesian information criterion values. Model parameter maps and model selection results were consistent with the physiology of the placenta and surrounding tissue. Conclusion Anisotropic intravoxel incoherent motion models explain the placental diffusion signal better than apparent diffusion coefficient, intravoxel incoherent motion, and diffusion tensor models, in information theoretic terms, when using this protocol. Future work will aim to determine if model‐derived parameters are sensitive to placental pathologies associated with disorders, such as fetal growth restriction and early‐onset pre‐eclampsia. Magn Reson Med 80:756–766, 2018. © 2017 The Authors Magnetic Resonance in Medicine published by Wiley Periodicals, Inc. on behalf of International Society for Magnetic Resonance in Medicine. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

As Supporting Table S2, but for the uterine wall ROI.

/ 10
Submitted to Magnetic Resonance in Medicine Supporting Figure S1: Parameter maps derived from DTI and ball-ball model fits.
Each row displays maps for a single slice from one subject, labelled by GA. Slices are displayed in the EPI acquisition plane, corresponding to the coronal plane (row 1 and 3) and axial plane (remaining rows). Arrows in row 7 highlight areas of high diffusivity and high perfusion at the boundary of the placenta.

/ 10
Submitted to Magnetic Resonance in Medicine Supporting Figure S2: Stick-zeppelin and zeppelin-zeppelin are close to the best model in most voxels.
Cumulative histograms of the difference between stick-zeppelin and zeppelin-zeppelin BICs, and the lowest BIC across all models in that voxel. A) Placenta ROI, B) uterine wall ROI.

/ 10
Submitted to Magnetic Resonance in Medicine Supporting Figure S3: Mapping the spatial pattern of model selection results.
Each row displays three slices for a single subject, labelled by GA. Voxels are coloured according to the category of the model with the lowest BIC in that voxel. Models are labelled according to the isotropy of the perfusion and diffusion compartments respectively, for example "aniso-iso" refers to models with anisotropic perfusion compartment and isotropic diffusion compartment. Slices are displayed in the EPI acquisition plane (coronal plane for rows 1 and 3, axial plane for other rows).

/ 10
Submitted to Magnetic Resonance in Medicine Supporting Figure S4: Parameter maps derived from stick-zeppelin model.
Each row displays maps for a single axial slice from one subject, labelled by GA. Slices are displayed in the EPI acquisition plane (coronal plane for rows 1 and 3, axial plane for other rows).

/ 10
Submitted to Magnetic Resonance in Medicine Supporting Figure S5: Standard deviation of stick-zeppelin parameters from bootstrapping analysis.
The data (i.e. 59 diffusion-weighted images) was resampled with replacement 100 times, and the stick-zeppelin model was fit to each resampled dataset. This enabled estimation of the standard deviation of stick-zeppelin model parameters (note that the color scales are 5 times lower than those in Figure 6 and Supporting Figure S4). Each row displays maps for a single axial slice from one subject, labelled by GA. Slices are displayed in the EPI acquisition plane (coronal plane for rows 1 and 3, axial plane for other rows).  As Supporting Figure S6 except plotting the median value of the perfusion fraction for three models.

/ 10
Submitted to Magnetic Resonance in Medicine stick-ball-sphere any other model Supporting Figure S8: Stick-ball-sphere parameter maps.
Each row displays maps for a single axial slice from one subject, labelled by GA. Slices are displayed in the acquisition plane. The second column shows the MD calculated from a DTI fit only to the images at b=0 and b=2000 s mm 2 . In the 5th row an area where stick-ball-sphere was the preferred model and the signal persisted at high b-values is circled, and arrows show areas with zero sphere volume fraction. In the 7th row an area with low sphere radius and non-zero sphere volume fraction is circled. 10 / 10