Implicit Modeling with Uncertainty Estimation for Intravoxel Incoherent Motion Imaging

TL;DR

This paper introduces an implicit neural network model for IVIM MRI that estimates perfusion parameters along with their uncertainty, significantly improving accuracy and repeatability over existing methods, especially in challenging fetal MRI conditions.

Contribution

The paper presents a novel neural network-based implicit modeling approach that estimates full posterior distributions of IVIM parameters, incorporating uncertainty quantification.

Findings

Improves IVIM parameter estimation accuracy by 65% on synthetic data.

Increases repeatability of fetal MRI placenta measurements by 46%.

Outperforms segmented least-squares fitting in accuracy.

Abstract

Intravoxel incoherent motion (IVIM) imaging allows contrast-agent free in vivo perfusion quantification with magnetic resonance imaging (MRI). However, its use is limited by typically low accuracy due to low signal-to-noise ratio (SNR) at large gradient encoding magnitudes as well as dephasing artefacts caused by subject motion, which is particularly challenging in fetal MRI. To mitigate this problem, we propose an implicit IVIM signal acquisition model with which we learn full posterior distribution of perfusion parameters using artificial neural networks. This posterior then encapsulates the uncertainty of the inferred parameter estimates, which we validate herein via numerical experiments with rejection-based Bayesian sampling. Compared to state-of-the-art IVIM estimation method of segmented least-squares fitting, our proposed approach improves parameter estimation accuracy by 65% on synthetic anisotropic perfusion data. On paired rescans of in vivo fetal MRI, our method increases repeatability of parameter estimation in placenta by 46%.