Continuous diffusion spectrum computation for diffusion-weighted magnetic resonance imaging of the kidney tubule system

Abstract

Background: The use of rigid multi-exponential models (with a priori predefined numbers of components) is common practice for diffusion-weighted MRI (DWI) analysis of the kidney. This approach may not accurately reflect renal microstructure, as the data are forced to conform to the a priori assumptions of simplified models. This work examines the feasibility of less constrained, data-driven non-negative least squares (NNLS) continuum modelling for DWI of the kidney tubule system in simulations that include emulations of pathophysiological conditions.Methods: Non-linear least squares (LS) fitting was used as reference for the simulations. For performance assessment, a threshold of 5% or 10% for the mean absolute percentage error (MAPE) of NNLS and LS results was used. As ground truth, a tri-exponential model using defined volume fractions and diffusion coefficients for each renal compartment (tubule system: D tubules , f tubules ; renal tissue: D tissue , f tissue ; renal blood: D blood , f blood ;) was applied. The impact of: (I) signal-to-noise ratio (SNR) =40-1,000, (II) number of b-values (n=10-50), (III) diffusion weighting (b-range small =0-800 up to b-range large =0-2,180 s/mm 2 ), and (IV) fixation of the diffusion coefficients D tissue and D blood was examined. NNLS was evaluated for baseline and pathophysiological conditions, namely increased tubular volume fraction (ITV) and renal fibrosis (10%: grade I, mild) and 30% (grade II, moderate).Results: NNLS showed the same high degree of reliability as the non-linear LS. MAPE of the tubular volume fraction (f tubules ) decreased with increasing SNR. Increasing the number of b-values was beneficial for f tubules precision. Using the b-range large led to a decrease in MAPE ftubules compared to b-range small . The use of a medium b-value range of b=0-1,380 s/mm 2 improved f tubules precision, and further b max increases beyond this range yielded diminishing improvements. Fixing D blood and D tissue significantly reduced MAPE ftubules and provided near perfect distinction between baseline and ITV conditions. Without constraining the number of renal compartments in advance, NNLS was able to detect the (fourth) fibrotic compartment, to differentiate it from the other three diffusion components, and to distinguish between 10% vs. 30% fibrosis.Conclusions: This work demonstrates the feasibility of NNLS modelling for DWI of the kidney tubule

Monitoring kidney size to interpret MRI‐based assessment of renal oxygenation in acute pathophysiological scenarios

“…Furthermore, advanced MR methodology supports monitoring of acute changes in the tubular or in the blood volume fraction using diffusion-weighted imaging. 60,61 The quantitative correction factors obtained for T 2 * and T 2 (slope and intercept of the linear regressions with KS) for the specific interventions are only valid for the present experimental setting. Although similar qualitative relationships between acute changes in T 2 *, T 2 and KS will exist for comparable acute interventions in preclinical and clinical studies, specific quantitative correction factors will naturally depend on the particular experimental or clinical setting including the magnetic field strength, and species.…”

Image downsampling expedited adaptive least‐squares (IDEAL) fitting improves intravoxel incoherent motion (IVIM) analysis in the human kidney

“…Consistent with the morphological and physiological characteristics of the kidney, evidence of the existence of an additional intermediate pseudodiffusion component in the renal tissue has been provided in a few recent studies in which less‐constrained, data‐driven nonnegative least‐squares continuum modelling has been utilized to determine the number of diffusion compartments present in the renal tissue 25,48 . Although the nonnegative least‐squares approach has the potential to reflect the complex nature of water diffusion in the kidney more accurately than the rigid multi‐exponential models, it is also susceptible to low SNR, leading to biased estimates and noisy parameter maps 49 …”

Image Downsampling Expedited Adaptive Least-squares (IDEAL) fitting improves intravoxel incoherent motion (IVIM) analysis in the human kidney

“…Consistent with the morphological and physiological characteristics of the kidney, evidence of the existence of an additional intermediate pseudo-diffusion component in the renal tissue has been provided in few recent studies in which less constrained, data-driven non-negative least squares (NNLS) continuum modelling has been utilized to determine the number of diffusion compartments present in the renal tissue 25,48 . Although the NNLS approach has the potential to reflect the complex nature of water diffusion in the kidney more accurately than the rigid multi-exponential models, it is also susceptible to low SNR leading to biased estimates and noisy parameter maps 49 .…”