NeXtQSM -- A complete deep learning pipeline for data-consistent quantitative susceptibility mapping trained with hybrid data

TL;DR

NeXtQSM introduces a deep learning pipeline for QSM that jointly performs background correction and dipole inversion in a data-consistent, end-to-end manner, improving robustness and speed over previous methods.

Contribution

It presents a hybrid data generation method and a variational network for joint, data-consistent QSM processing, addressing limitations of prior deep learning approaches.

Findings

NeXtQSM outperforms previous deep learning QSM methods.

The pipeline is robust and fast.

It effectively integrates multiple processing steps into one model.

Abstract

Deep learning based Quantitative Susceptibility Mapping (QSM) has shown great potential in recent years, obtaining similar results to established non-learning approaches. Many current deep learning approaches are not data consistent, require in vivo training data or solve the QSM problem in consecutive steps resulting in the propagation of errors. Here we aim to overcome these limitations and developed a framework to solve the QSM processing steps jointly. We developed a new hybrid training data generation method that enables the end-to-end training for solving background field correction and dipole inversion in a data-consistent fashion using a variational network that combines the QSM model term and a learned regularizer. We demonstrate that NeXtQSM overcomes the limitations of previous deep learning methods. NeXtQSM offers a new deep learning based pipeline for computing quantitative susceptibility maps that integrates each processing step into the training and provides results that are robust and fast.