Assessment of uncertainty and calibration of voxel-wise supervised modeling in IVIM

Introduction Intravoxel Incoherent Motion (IVIM) is a diffusion weighted (DW) MRI technique that models signal decay from tissue diffusion and capillary blood flows. Deep Neural Networks (DNNs) have emerged as a powerful alternative to traditional fitting methods, offering fast and robust inference with respect to conventional non-linear least squares (LSQ) and Bayesian approaches. DNNs typically provide only point estimates without uncertainty, which can help in understanding the impact of noise and guiding experimental design. This study, funded by a PRIN 2022 project, introduces a set of methods and metrics, based on the use of Mixture Density Networks (MDNs) [6] with Deep Ensembles (DEs) enabling, in a supervised voxel-wise setting, the estimation of both aleatoric (AU) and epistemic uncertainty (EU) and model calibration in IVIM.