Tri- and bi-exponential diffusion analyses of the kidney: effect of respiratory-controlled acquisition on diffusion...

This study examined whether respiratory-controlled acquisition influences diffusion parameters obtained with intravoxel incoherent motion (IVIM) analysis using tri-exponential and bi-exponential models. Ten healthy volunteers were examined on a 3.0 T MRI system to obtain coronal diffusion-weighted images of both kidneys. The participants were scanned twice using respiratory-triggering (RT) and free-breathing (FB) acquisition to assess the repeatability of the measurements. We determined mean signal intensities in the renal cortex at each b value. Then, perfusion-related diffusion coefficient (D_p), fast-free diffusion coefficient (D_f), slow-restricted diffusion coefficient (D_s), and their corresponding fractions (F_p, F_f, and F_s, respectively) were calculated using tri-exponential function. Moreover, perfusion-related diffusion coefficient (D*), the fraction (F), and perfusion-independent diffusion coefficient (D) were calculated using bi-exponential function. Normalized root-mean-square errors for the tri- and bi-exponential analyses (nRMSE_tri and nRMSE_bi, respectively) were determined to assess the deviation of the fitted to measured data, i.e., the fitting accuracy. Additionally, repeatability coefficients (RCs) were calculated from Bland-Altman plots to evaluate the repeatability of each diffusion parameter. These values were compared between the RT and FB groups. D_p and D* in the RT group were significantly lower than those in the FB group (P < 0.05). In addition, the RT group showed significantly lower nRMSE_tri and nRMSE_bi values than those in the FB group (P < 0.05). Moreover, D_p, D_s, F_s, and D* at RT showed lower RC values than those at FB. Respiratory-controlled acquisition affects perfusion-related diffusion parameters of the kidney obtained using tri-exponential and bi-exponential analyses.