BFRnet: A deep learning-based MR background field removal method for QSM of the brain containing significant pathological susceptibility sources

TL;DR

BFRnet is a deep learning method that effectively removes background magnetic field effects in brain QSM imaging, especially in cases with significant hemorrhagic sources, outperforming traditional methods in accuracy and robustness.

Contribution

This paper introduces BFRnet, a novel deep learning-based background field removal technique for QSM that handles pathological susceptibility sources and oblique acquisition orientations.

Findings

BFRnet achieved superior visual and quantitative results in local field and susceptibility maps.

BFRnet demonstrated robustness to brain mask variations and oblique FOV orientations.

Compared to conventional methods, BFRnet required less precise brain masking and edge erosion.

Abstract

Introduction: Background field removal (BFR) is a critical step required for successful quantitative susceptibility mapping (QSM). However, eliminating the background field in brains containing significant susceptibility sources, such as intracranial hemorrhages, is challenging due to the relatively large scale of the field induced by these pathological susceptibility sources. Method: This study proposes a new deep learning-based method, BFRnet, to remove background field in healthy and hemorrhagic subjects. The network is built with the dual-frequency octave convolutions on the U-net architecture, trained with synthetic field maps containing significant susceptibility sources. The BFRnet method is compared with three conventional BFR methods and one previous deep learning method using simulated and in vivo brains from 4 healthy and 2 hemorrhagic subjects. Robustness against acquisition field-of-view (FOV) orientation and brain masking are also investigated. Results: For both simulation and in vivo experiments, BFRnet led to the best visually appealing results in the local field and QSM results with the minimum contrast loss and the most accurate hemorrhage susceptibility measurements among all five methods. In addition, BFRnet produced the most consistent local field and susceptibility maps between different sizes of brain masks, while conventional methods depend drastically on precise brain extraction and further brain edge erosions. It is also observed that BFRnet performed the best among all BFR methods for acquisition FOVs oblique to the main magnetic field. Conclusion: The proposed BFRnet improved the accuracy of local field reconstruction in the hemorrhagic subjects compared with conventional BFR algorithms. The BFRnet method was effective for acquisitions of titled orientations and retained whole brains without edge erosion as often required by traditional BFR methods.