Overview of quantitative susceptibility mapping using deep learning: Current status, challenges and opportunities

Abstract

Quantitative susceptibility mapping (QSM) has gained broad interests in the field by extracting biological tissue properties, predominantly myelin, iron and calcium from magnetic resonance imaging (MRI) phase measurements in vivo. Thereby, QSM can reveal pathological changes of these key components in a variety of diseases. QSM requires multiple processing steps such as phase unwrapping, background field removal and field-to-source-inversion. Current state of the art techniques utilize iterative optimization procedures to solve the inversion and background field correction, which are computationally expensive and require a careful choice of regularization parameters. With the recent success of deep learning using convolutional neural networks for solving ill-posed reconstruction problems, the QSM community also adapted these techniques and demonstrated that the QSM processing steps can be solved by efficient feed forward multiplications not requiring iterative optimization nor the choice of regularization parameters.Here, we review the current status of deep learning based approaches for processing QSM, highlighting limitations and potential pitfalls, and discuss the future directions the field may take to exploit the latest advances in deep learning for QSM.

xQSM: quantitative susceptibility mapping with octave convolutional and noise‐regularized neural networks

“…By contrast, another deep learning framework, DeepQSM, 34 also based on the original U‐net, generated local field maps from synthetic susceptibility labels of simple geometric shapes using the forward model so that the training inputs and labels satisfied the exact underpinning equation between susceptibility source and induced field. However, the QSM results degraded noticeably when the test data deviated from the model (eg, noise and error in measurements) in DeepQSM 34,35 and susceptibility underestimation was observed, especially in deep gray matter (DGM) regions.…”

Electrical Properties Tomography: A Methodological Review

“…A key question is the generalization to data not encountered in the training. Interestingly, also in EPT's sister field of Quantitative Susceptibility Mapping deep learning improves the quality of susceptibility reconstruction over conventional methodology [103,104]. Promising results have been achieved here on the generalization issue.…”

QSM Reconstruction Challenge 2.0: a realistic in silico head phantom for MRI data simulation and evaluation of susceptibility mapping procedures

“…However, several recent methodical investigations have suggested that study outcomes may depend on the particular processing algorithms chosen for QSM (12)(13)(14). QSM typically involves the following steps in the order of application: coil combination (12); phase unwrapping (14); multi-echo combination (12); background field removal (14); and, finally, the estimation of susceptibility maps (13,(15)(16)(17). Processing artifacts and inaccuracies at any of these five processing stages can propagate into the computed susceptibility maps.…”