Nonlinear Dipole Inversion (NDI) enables Quantitative Susceptibility Mapping (QSM) without parameter tuning

TL;DR

The paper introduces Nonlinear Dipole Inversion (NDI), a novel method for quantitative susceptibility mapping that eliminates parameter tuning, improves reconstruction quality, and performs well with limited head orientation data.

Contribution

NDI combines a nonlinear forward model with deep learning to achieve high-quality QSM without regularization tuning or extensive parameter selection.

Findings

NDI matches state-of-the-art image quality without regularization tuning.

NDI outperforms COSMOS with as few as 1-direction data.

High-quality QSM achieved from 2-direction data at 7T using accelerated acquisitions.

Abstract

We propose Nonlinear Dipole Inversion (NDI) for high-quality Quantitative Susceptibility Mapping (QSM) without regularization tuning, while matching the image quality of state-of-the-art reconstruction techniques. In addition to avoiding over-smoothing that these techniques often suffer from, we also obviate the need for parameter selection. NDI is flexible enough to allow for reconstruction from an arbitrary number of head orientations, and outperforms COSMOS even when using as few as 1-direction data. This is made possible by a nonlinear forward-model that uses the magnitude as an effective prior, for which we derived a simple gradient descent update rule. We synergistically combine this physics-model with a Variational Network (VN) to leverage the power of deep learning in the VaNDI algorithm. This technique adopts the simple gradient descent rule from NDI and learns the network parameters during training, hence requires no additional parameter tuning. Further, we evaluate NDI at 7T using highly accelerated Wave-CAIPI acquisitions at 0.5 mm isotropic resolution and demonstrate high-quality QSM from as few as 2-direction data.